Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+examples/image2.jpg filter=lfs diff=lfs merge=lfs -text
+examples/red-panda.mp4 filter=lfs diff=lfs merge=lfs -text

NGC-DL-CONTAINER-LICENSE

@@ -0,0 +1,285 @@
+NVIDIA DEEP LEARNING CONTAINER LICENSE
+
+This license is a legal agreement between you and NVIDIA Corporation ("NVIDIA")
+and governs the use of the NVIDIA container and all its contents (“CONTAINER”).
+
+This license can be accepted only by an adult of legal age of majority in the
+country in which the CONTAINER is used. If you are under the legal age of
+majority, you must ask your parent or legal guardian to consent to this license.
+If you are entering this license on behalf of a company or other legal entity,
+you represent that you have legal authority and “you” will mean the entity you
+represent.
+
+By using the CONTAINER, you affirm that you have reached the legal age of
+majority, you accept the terms of this license, and you take legal and financial
+responsibility for the actions of your permitted users.
+
+You agree to use the CONTAINER only for purposes that are permitted by (a) this
+license, and (b) any applicable law, regulation or generally accepted practices
+or guidelines in the relevant jurisdictions.
+
+1. LICENSE. Subject to the terms of this license, NVIDIA hereby grants you a
+non-exclusive, non-transferable license, without the right to sublicense (except
+as expressly provided in this license) to:
+
+a. Install and use copies of the CONTAINER, and modify and create derivative
+works of samples or example source code delivered in the CONTAINER (if
+applicable), to develop and test services and applications,
+
+b. Deploy the CONTAINER on infrastructure you own or lease to offer a service to
+third parties, without distributing the CONTAINER or exposing the NVIDIA APIs in
+the CONTAINER directly to such service users, and
+
+c. Develop and extend the CONTAINER to create a Compatible (as defined below)
+derived CONTAINER that includes the entire CONTAINER plus other software with
+primary functionality, to develop and compile applications, and distribute such
+derived CONTAINER to run applications, subject to the distribution requirements
+indicated in this license. As used in this section, “Compatible” means that
+extensions to the CONTAINER must not adversely affect the functionality of the
+other components in the CONTAINER.
+
+2. DISTRIBUTION REQUIREMENTS. For purposes of this Section 2, the term
+“distribution” also means the deployment of CONTAINERS in a service or an
+application for third parties to access over the internet. These are the
+distribution requirements for you to exercise the grants above:
+
+a. A service or an application must have material additional functionality,
+beyond the included portions of the CONTAINER.
+
+b. The following notice shall be included in modifications and derivative works
+of source code distributed: “This software contains source code provided by
+NVIDIA Corporation.”
+
+c. You agree to distribute the CONTAINER subject to the terms at least as
+protective as the terms of this license, including (without limitation) terms
+relating to the license grant, license restrictions and protection of NVIDIA’s
+intellectual property rights. Additionally, you agree that you will protect the
+privacy, security and legal rights of your application users.
+
+d. You agree to notify NVIDIA in writing of any known or suspected distribution
+or use of the CONTAINER not in compliance with the requirements of this license,
+and to enforce the terms of your agreements with respect to the distributed
+CONTAINER.
+
+3. AUTHORIZED USERS. You may allow employees and contractors of your entity or
+of your subsidiary(ies) to access and use the CONTAINER from your secure network
+to perform work on your behalf. If you are an academic institution you may allow
+users enrolled or employed by the academic institution to access and use the
+CONTAINER from your secure network. You are responsible for the compliance with
+the terms of this license by your authorized users.
+
+4. LIMITATIONS. Your license to use the CONTAINER is restricted as follows:
+
+a. The CONTAINER may run on any computing system with or without NVIDIA GPUs,
+except for the NVIDIA proprietary software (such as CUDA and TensorRT software)
+in the CONTAINER which is licensed only to run on systems with NVIDIA GPUs. The
+NVIDIA proprietary software in the CONTAINER may be present on systems without
+NVIDIA GPUs, as long as it is not running on such systems. For components
+governed by open source software licenses, see the information in the
+“Components Under Other Licenses” section below.
+
+b. You may not reverse engineer, decompile or disassemble, or remove copyright
+or other proprietary notices from any portion of the CONTAINER or copies of the
+CONTAINER.
+
+c. Except as expressly provided in this license, you may not copy, sell, rent,
+sublicense, transfer, distribute, modify, or create derivative works of any
+portion of the CONTAINER. For clarity, you may not distribute or sublicense the
+CONTAINER as a stand-alone product.
+
+d. Unless you have an agreement with NVIDIA for this purpose, you may not
+indicate that a service or an application created with the CONTAINER is
+sponsored or endorsed by NVIDIA.
+
+e. You may not bypass, disable, or circumvent any technical limitation,
+encryption, security, digital rights management or authentication mechanism in
+the CONTAINER.
+
+f. You may not replace any NVIDIA software components in the CONTAINER that are
+governed by this license with other software that implements NVIDIA APIs.
+
+g. You may not use the CONTAINER in any manner that would cause it to become
+subject to an open source software license. As examples, licenses that require
+as a condition of use, modification, and/or distribution that the CONTAINER be:
+(i) disclosed or distributed in source code form; (ii) licensed for the purpose
+of making derivative works; or (iii) redistributable at no charge.
+
+h. You acknowledge that the CONTAINER as delivered is not tested or certified by
+NVIDIA for use in connection with the design, construction, maintenance, and/or
+operation of any system where the use or failure of such system could result in
+a situation that threatens the safety of human life or results in catastrophic
+damages (each, a “Critical Application”). Examples of Critical Applications
+include use in avionics, navigation, autonomous vehicle applications, ai
+solutions for automotive products, military, medical, life support or other life
+critical applications. NVIDIA shall not be liable to you or any third party, in
+whole or in part, for any claims or damages arising from such uses. You are
+solely responsible for ensuring that any product or service developed with the
+CONTAINER as a whole includes sufficient features to comply with all applicable
+legal and regulatory standards and requirements.
+
+i. You agree to defend, indemnify and hold harmless NVIDIA and its affiliates,
+and their respective employees, contractors, agents, officers and directors,
+from and against any and all claims, damages, obligations, losses, liabilities,
+costs or debt, fines, restitutions and expenses (including but not limited to
+attorney’s fees and costs incident to establishing the right of indemnification)
+arising out of or related to products or services that use the CONTAINER in or
+for Critical Applications, and for use of the CONTAINER outside of the scope of
+this license or not in compliance with its terms.
+
+j. You may not reverse engineer, decompile or disassemble any portion of the
+output generated using the NVIDIA proprietary software (such as CUDA and
+TensorRT software) in the CONTAINER for the purpose of translating such output
+artifacts to target a non-NVIDIA platform.
+
+5. UPDATES. NVIDIA may, at its option, make available patches, workarounds or
+other updates to this CONTAINER. Unless the updates are provided with their
+separate governing terms, they are deemed part of the CONTAINER licensed to you
+as provided in this license. You agree that the form and content of the
+CONTAINER that NVIDIA provides may change without prior notice to you. While
+NVIDIA generally maintains compatibility between versions, NVIDIA may in some
+cases make changes that introduce incompatibilities in future versions of the
+CONTAINER.
+
+6. PRE-RELEASE VERSIONS. CONTAINER versions identified as alpha, beta, preview,
+early access or otherwise as pre-release may not be fully functional, may
+contain errors or design flaws, and may have reduced or different security,
+privacy, availability, and reliability standards relative to commercial versions
+of NVIDIA software and materials. You may use a pre- release CONTAINER version
+at your own risk, understanding that these versions are not intended for use in
+production or business-critical systems. NVIDIA may choose not to make available
+a commercial version of any pre-release CONTAINER. NVIDIA may also choose to
+abandon development and terminate the availability of a pre-release CONTAINER at
+any time without liability.
+
+7. COMPONENTS UNDER OTHER LICENSES. The CONTAINER may include NVIDIA or
+third-party components with separate legal notices or terms as may be described
+in proprietary notices accompanying the CONTAINER. If and to the extent there is
+a conflict between the terms in this license and the license terms associated
+with the component, the license terms associated with the components control
+only to the extent necessary to resolve the conflict. For example, some Triton
+components (such as the Triton Inference Server) are governed by open source
+software licenses.
+
+You acknowledge and agree that it is your sole responsibility to obtain any
+additional third-party licenses required to make, have made, use, have used,
+sell, import, and offer for sale your products or services that include or
+incorporate any third- party software and content relating to audio and/or video
+encoders and decoders from, including but not limited to, Microsoft, Thomson,
+Fraunhofer IIS, Sisvel S.p.A., MPEG-LA, and Coding Technologies. NVIDIA does not
+grant to you under this license any necessary patent or other rights with
+respect to any audio and/or video encoders and decoders. Subject to the other
+terms of this license, you may use the CONTAINER to develop and test
+applications released under Open Source Initiative (OSI) approved open source
+software licenses.
+
+8. OWNERSHIP.
+
+8.1 NVIDIA reserves all rights, title and interest in and to the CONTAINER not
+expressly granted to you under this license. NVIDIA and its suppliers hold all
+rights, title and interest in and to the CONTAINER, including their respective
+intellectual property rights. The CONTAINER is copyrighted and protected by the
+laws of the United States and other countries, and international treaty
+provisions.
+
+8.2 Subject to the rights of NVIDIA and its suppliers in the CONTAINER, you hold
+all rights, title and interest in and to your services, applications and your
+derivative works of the sample source code delivered in the CONTAINER including
+their respective intellectual property rights.
+
+9. FEEDBACK. You may, but are not obligated to, provide to NVIDIA suggestions,
+fixes, modifications, feature requests or other feedback regarding the CONTAINER
+(“Feedback”). Feedback, even if designated as confidential by you, shall not
+create any confidentiality obligation for NVIDIA. NVIDIA and its designees have
+a perpetual, non-exclusive, worldwide, irrevocable license to use, reproduce,
+publicly display, modify, create derivative works of, license, sublicense, and
+otherwise distribute and exploit Feedback as NVIDIA sees fit without payment and
+without obligation or restriction of any kind on account of intellectual
+property rights or otherwise.
+
+10. NO WARRANTIES. THE CONTAINER IS PROVIDED AS-IS. TO THE MAXIMUM EXTENT
+PERMITTED BY APPLICABLE LAW NVIDIA AND ITS AFFILIATES EXPRESSLY DISCLAIM ALL
+WARRANTIES OF ANY KIND OR NATURE, WHETHER EXPRESS, IMPLIED OR STATUTORY,
+INCLUDING, BUT NOT LIMITED TO, WARRANTIES OF MERCHANTABILITY, NON-INFRINGEMENT,
+OR FITNESS FOR A PARTICULAR PURPOSE. NVIDIA DOES NOT WARRANT THAT THE CONTAINER
+WILL MEET YOUR REQUIREMENTS OR THAT THE OPERATION THEREOF WILL BE UNINTERRUPTED
+OR ERROR-FREE, OR THAT ALL ERRORS WILL BE CORRECTED.
+
+11. LIMITATIONS OF LIABILITY. TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW
+NVIDIA AND ITS AFFILIATES SHALL NOT BE LIABLE FOR ANY SPECIAL, INCIDENTAL,
+PUNITIVE OR CONSEQUENTIAL DAMAGES, OR FOR ANY LOST PROFITS, PROJECT DELAYS, LOSS
+OF USE, LOSS OF DATA OR LOSS OF GOODWILL, OR THE COSTS OF PROCURING SUBSTITUTE
+PRODUCTS, ARISING OUT OF OR IN CONNECTION WITH THIS LICENSE OR THE USE OR
+PERFORMANCE OF THE CONTAINER, WHETHER SUCH LIABILITY ARISES FROM ANY CLAIM BASED
+UPON BREACH OF CONTRACT, BREACH OF WARRANTY, TORT (INCLUDING NEGLIGENCE),
+PRODUCT LIABILITY OR ANY OTHER CAUSE OF ACTION OR THEORY OF LIABILITY, EVEN IF
+NVIDIA HAS PREVIOUSLY BEEN ADVISED OF, OR COULD REASONABLY HAVE FORESEEN, THE
+POSSIBILITY OF SUCH DAMAGES. IN NO EVENT WILL NVIDIA’S AND ITS AFFILIATES TOTAL
+CUMULATIVE LIABILITY UNDER OR ARISING OUT OF THIS LICENSE EXCEED US$10.00. THE
+NATURE OF THE LIABILITY OR THE NUMBER OF CLAIMS OR SUITS SHALL NOT ENLARGE OR
+EXTEND THIS LIMIT.
+
+12. TERMINATION. Your rights under this license will terminate automatically
+without notice from NVIDIA if you fail to comply with any term and condition of
+this license or if you commence or participate in any legal proceeding against
+NVIDIA with respect to the CONTAINER. NVIDIA may terminate this license with
+advance written notice to you, if NVIDIA decides to no longer provide the
+CONTAINER in a country or, in NVIDIA’s sole discretion, the continued use of it
+is no longer commercially viable. Upon any termination of this license, you
+agree to promptly discontinue use of the CONTAINER and destroy all copies in
+your possession or control. Your prior distributions in accordance with this
+license are not affected by the termination of this license. All provisions of
+this license will survive termination, except for the license granted to you.
+
+13. APPLICABLE LAW. This license will be governed in all respects by the laws of
+the United States and of the State of Delaware, without regard to the conflicts
+of laws principles. The United Nations Convention on Contracts for the
+International Sale of Goods is specifically disclaimed. You agree to all terms
+of this license in the English language. The state or federal courts residing in
+Santa Clara County, California shall have exclusive jurisdiction over any
+dispute or claim arising out of this license. Notwithstanding this, you agree
+that NVIDIA shall still be allowed to apply for injunctive remedies or urgent
+legal relief in any jurisdiction.
+
+14. NO ASSIGNMENT. This license and your rights and obligations thereunder may
+not be assigned by you by any means or operation of law without NVIDIA’s
+permission. Any attempted assignment not approved by NVIDIA in writing shall be
+void and of no effect. NVIDIA may assign, delegate or transfer this license and
+its rights and obligations, and if to a non-affiliate you will be notified.
+
+15. EXPORT. The CONTAINER is subject to United States export laws and
+regulations. You agree to comply with all applicable U.S. and international
+export laws, including the Export Administration Regulations (EAR) administered
+by the U.S. Department of Commerce and economic sanctions administered by the
+U.S. Department of Treasury’s Office of Foreign Assets Control (OFAC). These
+laws include restrictions on destinations, end-users and end-use. By accepting
+this license, you confirm that you are not currently residing in a country or
+region currently embargoed by the U.S. and that you are not otherwise prohibited
+from receiving the CONTAINER.
+
+16. GOVERNMENT USE. The CONTAINER is, and shall be treated as being, “Commercial
+Items” as that term is defined at 48 CFR § 2.101, consisting of “commercial
+computer software” and “commercial computer software documentation”,
+respectively, as such terms are used in, respectively, 48 CFR § 12.212 and 48
+CFR §§ 227.7202 & 252.227-7014(a)(1). Use, duplication or disclosure by the U.S.
+Government or a U.S. Government subcontractor is subject to the restrictions in
+this license pursuant to 48 CFR § 12.212 or 48 CFR § 227.7202. In no event shall
+the US Government user acquire rights in the CONTAINER beyond those specified in
+48 C.F.R. 52.227-19(b)(1)-(2).
+
+17. NOTICES. Please direct your legal notices or other correspondence to NVIDIA
+Corporation, 2788 San Tomas Expressway, Santa Clara, California 95051, United
+States of America, Attention: Legal Department. 18. ENTIRE AGREEMENT. This
+license is the final, complete and exclusive agreement between the parties
+relating to the subject matter of this license and supersedes all prior or
+contemporaneous understandings and agreements relating to this subject matter,
+whether oral or written. If any court of competent jurisdiction determines that
+any provision of this license is illegal, invalid or unenforceable, the
+remaining provisions will remain in full force and effect. Any amendment or
+waiver under this license shall be in writing and signed by representatives of
+both parties.
+
+19. LICENSING. If the distribution terms in this license are not suitable for
+your organization, or for any questions regarding this license, please contact
+NVIDIA at [email protected].
+
+(v. September 14, 2021)

README.md

@@ -0,0 +1,703 @@
+---
+license: apache-2.0
+license_name: qwen
+license_link: https://huggingface.co/Qwen/Qwen2.5-72B-Instruct/blob/main/LICENSE
+pipeline_tag: image-text-to-text
+library_name: transformers
+base_model:
+- OpenGVLab/InternVL3-14B-Instruct
+base_model_relation: finetune
+datasets:
+- OpenGVLab/MMPR-v1.2
+language:
+- multilingual
+tags:
+- internvl
+- custom_code
+---
+
+# InternVL3-14B
+
+[\[📂 GitHub\]](https://github.com/OpenGVLab/InternVL)  [\[📜 InternVL 1.0\]](https://huggingface.co/papers/2312.14238)  [\[📜 InternVL 1.5\]](https://huggingface.co/papers/2404.16821)  [\[📜 InternVL 2.5\]](https://huggingface.co/papers/2412.05271)  [\[📜 InternVL2.5-MPO\]](https://huggingface.co/papers/2411.10442)  [\[📜 InternVL3\]](https://huggingface.co/papers/2504.10479)
+
+[\[🆕 Blog\]](https://internvl.github.io/blog/)  [\[🗨️ Chat Demo\]](https://internvl.opengvlab.com/)  [\[🤗 HF Demo\]](https://huggingface.co/spaces/OpenGVLab/InternVL)  [\[🚀 Quick Start\]](#quick-start)  [\[📖 Documents\]](https://internvl.readthedocs.io/en/latest/)
+
+<div align="center">
+  <img width="500" alt="image" src="https://cdn-uploads.huggingface.co/production/uploads/64006c09330a45b03605bba3/zJsd2hqd3EevgXo6fNgC-.png">
+</div>
+
+## Introduction
+
+We introduce InternVL3, an advanced multimodal large language model (MLLM) series that demonstrates superior overall performance.
+Compared to InternVL 2.5, InternVL3 exhibits superior multimodal perception and reasoning capabilities, while further extending its multimodal capabilities to encompass tool usage, GUI agents, industrial image analysis, 3D vision perception, and more.
+Additionally, we compare InternVL3 with  Qwen2.5 Chat models, whose corresponding pre-trained base models are employed as the initialization of the langauge component in InternVL3. Benefitting from Native Multimodal Pre-Training, the InternVL3 series achieves even better overall text performance than the Qwen2.5 series.
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/overall.png)
+
+## InternVL3 Family
+
+In the following table, we provide an overview of the InternVL3 series.
+
+|  Model Name   |                                       Vision Part                                       |                                 Language Part                                  |                         HF Link                          |
+| :-----------: | :-------------------------------------------------------------------------------------: | :----------------------------------------------------------------------------: | :------------------------------------------------------: |
+| InternVL3-1B  | [InternViT-300M-448px-V2_5](https://huggingface.co/OpenGVLab/InternViT-300M-448px-V2_5) |            [Qwen2.5-0.5B](https://huggingface.co/Qwen/Qwen2.5-0.5B)            | [🤗 link](https://huggingface.co/OpenGVLab/InternVL3-1B)  |
+| InternVL3-2B  | [InternViT-300M-448px-V2_5](https://huggingface.co/OpenGVLab/InternViT-300M-448px-V2_5) |            [Qwen2.5-1.5B](https://huggingface.co/Qwen/Qwen2.5-1.5B)            | [🤗 link](https://huggingface.co/OpenGVLab/InternVL3-2B)  |
+| InternVL3-8B  | [InternViT-300M-448px-V2_5](https://huggingface.co/OpenGVLab/InternViT-300M-448px-V2_5) |              [Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B)              | [🤗 link](https://huggingface.co/OpenGVLab/InternVL3-8B)  |
+| InternVL3-9B  | [InternViT-300M-448px-V2_5](https://huggingface.co/OpenGVLab/InternViT-300M-448px-V2_5) | [internlm3-8b-instruct](https://huggingface.co/internlm/internlm3-8b-instruct) | [🤗 link](https://huggingface.co/OpenGVLab/InternVL3-9B)  |
+| InternVL3-14B | [InternViT-300M-448px-V2_5](https://huggingface.co/OpenGVLab/InternViT-300M-448px-V2_5) |             [Qwen2.5-14B](https://huggingface.co/Qwen/Qwen2.5-14B)             | [🤗 link](https://huggingface.co/OpenGVLab/InternVL3-14B) |
+| InternVL3-38B |   [InternViT-6B-448px-V2_5](https://huggingface.co/OpenGVLab/InternViT-6B-448px-V2_5)   |             [Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B)             | [🤗 link](https://huggingface.co/OpenGVLab/InternVL3-38B) |
+| InternVL3-78B |   [InternViT-6B-448px-V2_5](https://huggingface.co/OpenGVLab/InternViT-6B-448px-V2_5)   |             [Qwen2.5-72B](https://huggingface.co/Qwen/Qwen2.5-72B)             | [🤗 link](https://huggingface.co/OpenGVLab/InternVL3-78B) |
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/overall-table.png)
+
+## Model Architecture
+
+As shown in the following figure, [InternVL3](https://internvl.github.io/blog/2025-04-11-InternVL-3/) retains the same model architecture as [InternVL 2.5](https://internvl.github.io/blog/2024-12-05-InternVL-2.5/) and its predecessors, InternVL 1.5 and 2.0, following the "ViT-MLP-LLM" paradigm. In this new version, we integrate a newly incrementally pre-trained InternViT with various pre-trained LLMs, including InternLM 3 and Qwen 2.5, using a randomly initialized MLP projector.
+
+
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/64119264f0f81eb569e0d569/BiiyXN6NOk0p-3rl3ueyL.png)
+
+As in the previous version, we applied a pixel unshuffle operation, reducing the number of visual tokens to one-quarter of the original. Besides, we adopted a similar dynamic resolution strategy as InternVL 1.5, dividing images into tiles of 448×448 pixels. The key difference, starting from InternVL 2.0, is that we additionally introduced support for multi-image and video data.
+
+Notably, in InternVL3, we integrate the [Variable Visual Position Encoding (V2PE)](https://arxiv.org/abs/2412.09616), which utilizes smaller, more flexible position increments for visual tokens. Benefiting from V2PE, InternVL3 exhibits better long context understanding capabilities compared to its predecessors.
+
+## Training Strategy
+
+### Native Multimodal Pre-Training
+
+We propose a [Native Multimodal Pre-Training](https://huggingface.co/papers/2504.10479) approach that consolidates language and vision learning into a single pre-training stage.
+In contrast to standard paradigms that first train a language-only model and subsequently adapt it to handle additional modalities, our method interleaves multimodal data (e.g., image-text, video-text, or image-text interleaved sequences) with large-scale textual corpora. This unified training scheme allows the model to learn both linguistic and multimodal representations simultaneously, ultimately enhancing its capability to handle vision-language tasks without the need for separate alignment or bridging modules.
+Please see [our paper](https://huggingface.co/papers/2504.10479) for more details.
+
+### Supervised Fine-Tuning
+
+In this phase, the techniques of random JPEG compression, square loss re-weighting, and multimodal data packing proposed in [InternVL2.5](https://arxiv.org/abs/2412.05271) are also employed in the InternVL3 series.
+The main advancement of the SFT phase in InternVL3 compared to InternVL2.5 lies in the use of higher-quality and more diverse training data.
+Specifically, we further extend  training samples for tool use, 3D scene understanding, GUI operations, long context tasks, video understanding, scientific diagrams, creative writing, and multimodal reasoning.
+
+### Mixed Preference Optimization
+
+During Pre-training and SFT, the model is trained to predict the next token conditioned on previous ground-truth tokens.
+However, during inference, the model predicts each token based on its own prior outputs. 
+This discrepancy between ground-truth tokens and model-predicted tokens introduces a distribution shift, which can impair the model’s Chain-of-Thought (CoT) reasoning capabilities.
+To mitigate this issue, we employ [MPO](https://arxiv.org/abs/2411.10442), which introduces additional supervision from both positive and negative samples to align the model response distribution with the ground-truth distribution, thereby improving reasoning performance.
+Specifically, the training objective of MPO is a combination of
+preference loss \\(\mathcal{L}_{\text{p}}\\),
+quality loss \\(\mathcal{L}_{\text{q}}\\),
+and generation loss \\(\mathcal{L}_{\text{g}}\\),
+which can be formulated as follows:
+
+
+$$
+\mathcal{L}=w_{p}\cdot\mathcal{L}_{\text{p}} + w_{q}\cdot\mathcal{L}_{\text{q}} + w_{g}\cdot\mathcal{L}_{\text{g}},
+$$
+
+
+where \\(w_{*}\\) represents the weight assigned to each loss component. Please see [our paper](https://arxiv.org/abs/2411.10442) for more details about MPO.
+
+
+### Test-Time Scaling
+
+Test-Time Scaling has been shown to be an effective method to enhance the reasoning abilities of LLMs and MLLMs.
+In this work, we use the Best-of-N evaluation strategy and employ [VisualPRM-8B](https://huggingface.co/OpenGVLab/VisualPRM-8B) as the critic model to select the best response for reasoning and mathematics evaluation.
+
+## Evaluation on Multimodal Capability
+
+### Multimodal Reasoning and Mathematics
+
+![image/png](https://huggingface.co/OpenGVLab/VisualPRM-8B-v1_1/resolve/main/visualprm-performance.png)
+
+### OCR, Chart, and Document Understanding
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/ocr.png)
+
+### Multi-Image & Real-World Comprehension
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/multi-images.png)
+
+### Comprehensive Multimodal & Hallucination Evaluation
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/comprehensive.png)
+
+### Visual Grounding
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/grounding.png)
+
+### Multimodal Multilingual Understanding
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/multilingual.png)
+
+### Video Understanding
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/video.png)
+
+### GUI Grounding
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/gui.png)
+
+### Spatial Reasoning
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/vsi.png)
+
+## Evaluation on Language Capability
+
+We compare InternVL3 with  Qwen2.5 Chat models, whose corresponding pre-trained base models are employed as the initialization of the langauge component in InternVL3.
+Benefitting from Native Multimodal Pre-Training, the InternVL3 series achieves even better overall text performance than the Qwen2.5 series.
+Please note that the evaluation scores of Qwen2.5 series  may differ from those officially reported, as we have adopted the prompt versions provided in the table across all datasets for OpenCompass evaluation.
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/text.png)
+
+## Ablation Study
+
+### Native Multimodal Pre-Training
+
+We conduct experiments on the InternVL2-8B model while keeping its architecture, initialization parameters, and training data entirely unchanged. Traditionally, InternVL2-8B employs a training pipeline that begins with an MLP warmup phase for feature alignment followed by an Instruction Tuning stage. In our experiments, we substitute the conventional MLP warmup phase with a native multimodal pre-training process. This modification isolates the contribution of native multimodal pre-training to the overall multimodal capability of the model.
+
+The evaluation results in the Figure below shows that the model with native multimodal pre-training exhibits performance on most benchmarks that is comparable to the fully multi-stage-trained InternVL2-8B baseline. Furthermore, when followed by instruction tuning on higher-quality data, the model demonstrates further performance gains across evaluated multimodal tasks. These findings underscore the efficiency of native multimodal pre-training in imparting powerful multimodal capabilities to MLLMs.
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/ablation-native.png)
+
+### Mixed Preference Optimization
+
+As shown in the table below, models fine-tuned with MPO demonstrate superior reasoning performance across seven multimodal reasoning benchmarks compared to their counterparts without MPO. Specifically, InternVL3-78B and InternVL3-38B outperform their counterparts by 4.1 and 4.5 points, respectively. Notably, the training data used for MPO is a subset of that used for SFT, indicating that the performance improvements primarily stem from the training algorithm rather than the training data.
+
+![image/png](https://huggingface.co/datasets/OpenGVLab/MMPR-v1.2-prompts/resolve/main/ablation-mpo.png)
+
+### Variable Visual Position Encoding
+
+As reported in the table below, the introduction of V2PE leads to significant performance gains across most evaluation metrics. In addition, our ablation studies—by varying the positional increment \\( \delta \\)—reveal that even for tasks primarily involving conventional contexts, relatively small \\( \delta \\) values can achieve optimal performance. These findings provide important insights for future efforts aimed at refining position encoding strategies for visual tokens in MLLMs.
+
+![image/png](https://huggingface.co/datasets/Weiyun1025/InternVL-Performance/resolve/main/internvl3/ablation-v2pe.png)
+
+## Quick Start
+
+We provide an example code to run `InternVL3-14B` using `transformers`.
+
+> Please use transformers>=4.37.2 to ensure the model works normally.
+
+### Model Loading
+
+#### 16-bit (bf16 / fp16)
+
+```python
+import torch
+from transformers import AutoTokenizer, AutoModel
+path = "OpenGVLab/InternVL3-14B"
+model = AutoModel.from_pretrained(
+    path,
+    torch_dtype=torch.bfloat16,
+    low_cpu_mem_usage=True,
+    use_flash_attn=True,
+    trust_remote_code=True).eval().cuda()
+```
+
+#### BNB 8-bit Quantization
+
+```python
+import torch
+from transformers import AutoTokenizer, AutoModel
+path = "OpenGVLab/InternVL3-14B"
+model = AutoModel.from_pretrained(
+    path,
+    torch_dtype=torch.bfloat16,
+    load_in_8bit=True,
+    low_cpu_mem_usage=True,
+    use_flash_attn=True,
+    trust_remote_code=True).eval()
+```
+
+#### Multiple GPUs
+
+The reason for writing the code this way is to avoid errors that occur during multi-GPU inference due to tensors not being on the same device. By ensuring that the first and last layers of the large language model (LLM) are on the same device, we prevent such errors.
+
+```python
+import math
+import torch
+from transformers import AutoTokenizer, AutoModel
+
+def split_model(model_name):
+    device_map = {}
+    world_size = torch.cuda.device_count()
+    config = AutoConfig.from_pretrained(model_path, trust_remote_code=True)
+    num_layers = config.llm_config.num_hidden_layers
+    # Since the first GPU will be used for ViT, treat it as half a GPU.
+    num_layers_per_gpu = math.ceil(num_layers / (world_size - 0.5))
+    num_layers_per_gpu = [num_layers_per_gpu] * world_size
+    num_layers_per_gpu[0] = math.ceil(num_layers_per_gpu[0] * 0.5)
+    layer_cnt = 0
+    for i, num_layer in enumerate(num_layers_per_gpu):
+        for j in range(num_layer):
+            device_map[f'language_model.model.layers.{layer_cnt}'] = i
+            layer_cnt += 1
+    device_map['vision_model'] = 0
+    device_map['mlp1'] = 0
+    device_map['language_model.model.tok_embeddings'] = 0
+    device_map['language_model.model.embed_tokens'] = 0
+    device_map['language_model.output'] = 0
+    device_map['language_model.model.norm'] = 0
+    device_map['language_model.model.rotary_emb'] = 0
+    device_map['language_model.lm_head'] = 0
+    device_map[f'language_model.model.layers.{num_layers - 1}'] = 0
+
+    return device_map
+
+path = "OpenGVLab/InternVL3-14B"
+device_map = split_model('InternVL3-14B')
+model = AutoModel.from_pretrained(
+    path,
+    torch_dtype=torch.bfloat16,
+    low_cpu_mem_usage=True,
+    use_flash_attn=True,
+    trust_remote_code=True,
+    device_map=device_map).eval()
+```
+
+### Inference with Transformers
+
+```python
+import math
+import numpy as np
+import torch
+import torchvision.transforms as T
+from decord import VideoReader, cpu
+from PIL import Image
+from torchvision.transforms.functional import InterpolationMode
+from transformers import AutoModel, AutoTokenizer
+
+IMAGENET_MEAN = (0.485, 0.456, 0.406)
+IMAGENET_STD = (0.229, 0.224, 0.225)
+
+def build_transform(input_size):
+    MEAN, STD = IMAGENET_MEAN, IMAGENET_STD
+    transform = T.Compose([
+        T.Lambda(lambda img: img.convert('RGB') if img.mode != 'RGB' else img),
+        T.Resize((input_size, input_size), interpolation=InterpolationMode.BICUBIC),
+        T.ToTensor(),
+        T.Normalize(mean=MEAN, std=STD)
+    ])
+    return transform
+
+def find_closest_aspect_ratio(aspect_ratio, target_ratios, width, height, image_size):
+    best_ratio_diff = float('inf')
+    best_ratio = (1, 1)
+    area = width * height
+    for ratio in target_ratios:
+        target_aspect_ratio = ratio[0] / ratio[1]
+        ratio_diff = abs(aspect_ratio - target_aspect_ratio)
+        if ratio_diff < best_ratio_diff:
+            best_ratio_diff = ratio_diff
+            best_ratio = ratio
+        elif ratio_diff == best_ratio_diff:
+            if area > 0.5 * image_size * image_size * ratio[0] * ratio[1]:
+                best_ratio = ratio
+    return best_ratio
+
+def dynamic_preprocess(image, min_num=1, max_num=12, image_size=448, use_thumbnail=False):
+    orig_width, orig_height = image.size
+    aspect_ratio = orig_width / orig_height
+
+    # calculate the existing image aspect ratio
+    target_ratios = set(
+        (i, j) for n in range(min_num, max_num + 1) for i in range(1, n + 1) for j in range(1, n + 1) if
+        i * j <= max_num and i * j >= min_num)
+    target_ratios = sorted(target_ratios, key=lambda x: x[0] * x[1])
+
+    # find the closest aspect ratio to the target
+    target_aspect_ratio = find_closest_aspect_ratio(
+        aspect_ratio, target_ratios, orig_width, orig_height, image_size)
+
+    # calculate the target width and height
+    target_width = image_size * target_aspect_ratio[0]
+    target_height = image_size * target_aspect_ratio[1]
+    blocks = target_aspect_ratio[0] * target_aspect_ratio[1]
+
+    # resize the image
+    resized_img = image.resize((target_width, target_height))
+    processed_images = []
+    for i in range(blocks):
+        box = (
+            (i % (target_width // image_size)) * image_size,
+            (i // (target_width // image_size)) * image_size,
+            ((i % (target_width // image_size)) + 1) * image_size,
+            ((i // (target_width // image_size)) + 1) * image_size
+        )
+        # split the image
+        split_img = resized_img.crop(box)
+        processed_images.append(split_img)
+    assert len(processed_images) == blocks
+    if use_thumbnail and len(processed_images) != 1:
+        thumbnail_img = image.resize((image_size, image_size))
+        processed_images.append(thumbnail_img)
+    return processed_images
+
+def load_image(image_file, input_size=448, max_num=12):
+    image = Image.open(image_file).convert('RGB')
+    transform = build_transform(input_size=input_size)
+    images = dynamic_preprocess(image, image_size=input_size, use_thumbnail=True, max_num=max_num)
+    pixel_values = [transform(image) for image in images]
+    pixel_values = torch.stack(pixel_values)
+    return pixel_values
+
+def split_model(model_name):
+    device_map = {}
+    world_size = torch.cuda.device_count()
+    config = AutoConfig.from_pretrained(model_path, trust_remote_code=True)
+    num_layers = config.llm_config.num_hidden_layers
+    # Since the first GPU will be used for ViT, treat it as half a GPU.
+    num_layers_per_gpu = math.ceil(num_layers / (world_size - 0.5))
+    num_layers_per_gpu = [num_layers_per_gpu] * world_size
+    num_layers_per_gpu[0] = math.ceil(num_layers_per_gpu[0] * 0.5)
+    layer_cnt = 0
+    for i, num_layer in enumerate(num_layers_per_gpu):
+        for j in range(num_layer):
+            device_map[f'language_model.model.layers.{layer_cnt}'] = i
+            layer_cnt += 1
+    device_map['vision_model'] = 0
+    device_map['mlp1'] = 0
+    device_map['language_model.model.tok_embeddings'] = 0
+    device_map['language_model.model.embed_tokens'] = 0
+    device_map['language_model.output'] = 0
+    device_map['language_model.model.norm'] = 0
+    device_map['language_model.model.rotary_emb'] = 0
+    device_map['language_model.lm_head'] = 0
+    device_map[f'language_model.model.layers.{num_layers - 1}'] = 0
+
+    return device_map
+
+# If you set `load_in_8bit=True`, you will need two 80GB GPUs.
+# If you set `load_in_8bit=False`, you will need at least three 80GB GPUs.
+path = 'OpenGVLab/InternVL3-14B'
+device_map = split_model('InternVL3-14B')
+model = AutoModel.from_pretrained(
+    path,
+    torch_dtype=torch.bfloat16,
+    load_in_8bit=False,
+    low_cpu_mem_usage=True,
+    use_flash_attn=True,
+    trust_remote_code=True,
+    device_map=device_map).eval()
+tokenizer = AutoTokenizer.from_pretrained(path, trust_remote_code=True, use_fast=False)
+
+# set the max number of tiles in `max_num`
+pixel_values = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
+generation_config = dict(max_new_tokens=1024, do_sample=True)
+
+# pure-text conversation (纯文本对话)
+question = 'Hello, who are you?'
+response, history = model.chat(tokenizer, None, question, generation_config, history=None, return_history=True)
+print(f'User: {question}\nAssistant: {response}')
+
+question = 'Can you tell me a story?'
+response, history = model.chat(tokenizer, None, question, generation_config, history=history, return_history=True)
+print(f'User: {question}\nAssistant: {response}')
+
+# single-image single-round conversation (单图单轮对话)
+question = '<image>\nPlease describe the image shortly.'
+response = model.chat(tokenizer, pixel_values, question, generation_config)
+print(f'User: {question}\nAssistant: {response}')
+
+# single-image multi-round conversation (单图多轮对话)
+question = '<image>\nPlease describe the image in detail.'
+response, history = model.chat(tokenizer, pixel_values, question, generation_config, history=None, return_history=True)
+print(f'User: {question}\nAssistant: {response}')
+
+question = 'Please write a poem according to the image.'
+response, history = model.chat(tokenizer, pixel_values, question, generation_config, history=history, return_history=True)
+print(f'User: {question}\nAssistant: {response}')
+
+# multi-image multi-round conversation, combined images (多图多轮对话，拼接图像)
+pixel_values1 = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
+pixel_values2 = load_image('./examples/image2.jpg', max_num=12).to(torch.bfloat16).cuda()
+pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)
+
+question = '<image>\nDescribe the two images in detail.'
+response, history = model.chat(tokenizer, pixel_values, question, generation_config,
+                               history=None, return_history=True)
+print(f'User: {question}\nAssistant: {response}')
+
+question = 'What are the similarities and differences between these two images.'
+response, history = model.chat(tokenizer, pixel_values, question, generation_config,
+                               history=history, return_history=True)
+print(f'User: {question}\nAssistant: {response}')
+
+# multi-image multi-round conversation, separate images (多图多轮对话，独立图像)
+pixel_values1 = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
+pixel_values2 = load_image('./examples/image2.jpg', max_num=12).to(torch.bfloat16).cuda()
+pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)
+num_patches_list = [pixel_values1.size(0), pixel_values2.size(0)]
+
+question = 'Image-1: <image>\nImage-2: <image>\nDescribe the two images in detail.'
+response, history = model.chat(tokenizer, pixel_values, question, generation_config,
+                               num_patches_list=num_patches_list,
+                               history=None, return_history=True)
+print(f'User: {question}\nAssistant: {response}')
+
+question = 'What are the similarities and differences between these two images.'
+response, history = model.chat(tokenizer, pixel_values, question, generation_config,
+                               num_patches_list=num_patches_list,
+                               history=history, return_history=True)
+print(f'User: {question}\nAssistant: {response}')
+
+# batch inference, single image per sample (单图批处理)
+pixel_values1 = load_image('./examples/image1.jpg', max_num=12).to(torch.bfloat16).cuda()
+pixel_values2 = load_image('./examples/image2.jpg', max_num=12).to(torch.bfloat16).cuda()
+num_patches_list = [pixel_values1.size(0), pixel_values2.size(0)]
+pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)
+
+questions = ['<image>\nDescribe the image in detail.'] * len(num_patches_list)
+responses = model.batch_chat(tokenizer, pixel_values,
+                             num_patches_list=num_patches_list,
+                             questions=questions,
+                             generation_config=generation_config)
+for question, response in zip(questions, responses):
+    print(f'User: {question}\nAssistant: {response}')
+
+# video multi-round conversation (视频多轮对话)
+def get_index(bound, fps, max_frame, first_idx=0, num_segments=32):
+    if bound:
+        start, end = bound[0], bound[1]
+    else:
+        start, end = -100000, 100000
+    start_idx = max(first_idx, round(start * fps))
+    end_idx = min(round(end * fps), max_frame)
+    seg_size = float(end_idx - start_idx) / num_segments
+    frame_indices = np.array([
+        int(start_idx + (seg_size / 2) + np.round(seg_size * idx))
+        for idx in range(num_segments)
+    ])
+    return frame_indices
+
+def load_video(video_path, bound=None, input_size=448, max_num=1, num_segments=32):
+    vr = VideoReader(video_path, ctx=cpu(0), num_threads=1)
+    max_frame = len(vr) - 1
+    fps = float(vr.get_avg_fps())
+
+    pixel_values_list, num_patches_list = [], []
+    transform = build_transform(input_size=input_size)
+    frame_indices = get_index(bound, fps, max_frame, first_idx=0, num_segments=num_segments)
+    for frame_index in frame_indices:
+        img = Image.fromarray(vr[frame_index].asnumpy()).convert('RGB')
+        img = dynamic_preprocess(img, image_size=input_size, use_thumbnail=True, max_num=max_num)
+        pixel_values = [transform(tile) for tile in img]
+        pixel_values = torch.stack(pixel_values)
+        num_patches_list.append(pixel_values.shape[0])
+        pixel_values_list.append(pixel_values)
+    pixel_values = torch.cat(pixel_values_list)
+    return pixel_values, num_patches_list
+
+video_path = './examples/red-panda.mp4'
+pixel_values, num_patches_list = load_video(video_path, num_segments=8, max_num=1)
+pixel_values = pixel_values.to(torch.bfloat16).cuda()
+video_prefix = ''.join([f'Frame{i+1}: <image>\n' for i in range(len(num_patches_list))])
+question = video_prefix + 'What is the red panda doing?'
+# Frame1: <image>\nFrame2: <image>\n...\nFrame8: <image>\n{question}
+response, history = model.chat(tokenizer, pixel_values, question, generation_config,
+                               num_patches_list=num_patches_list, history=None, return_history=True)
+print(f'User: {question}\nAssistant: {response}')
+
+question = 'Describe this video in detail.'
+response, history = model.chat(tokenizer, pixel_values, question, generation_config,
+                               num_patches_list=num_patches_list, history=history, return_history=True)
+print(f'User: {question}\nAssistant: {response}')
+```
+
+#### Streaming Output
+
+Besides this method, you can also use the following code to get streamed output.
+
+```python
+from transformers import TextIteratorStreamer
+from threading import Thread
+
+# Initialize the streamer
+streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True, timeout=10)
+# Define the generation configuration
+generation_config = dict(max_new_tokens=1024, do_sample=False, streamer=streamer)
+# Start the model chat in a separate thread
+thread = Thread(target=model.chat, kwargs=dict(
+    tokenizer=tokenizer, pixel_values=pixel_values, question=question,
+    history=None, return_history=False, generation_config=generation_config,
+))
+thread.start()
+
+# Initialize an empty string to store the generated text
+generated_text = ''
+# Loop through the streamer to get the new text as it is generated
+for new_text in streamer:
+    if new_text == model.conv_template.sep:
+        break
+    generated_text += new_text
+    print(new_text, end='', flush=True)  # Print each new chunk of generated text on the same line
+```
+
+## Finetune
+
+Many repositories now support fine-tuning of the InternVL series models, including [InternVL](https://github.com/OpenGVLab/InternVL), [SWIFT](https://github.com/modelscope/ms-swift), [XTurner](https://github.com/InternLM/xtuner), and others. Please refer to their documentation for more details on fine-tuning.
+
+## Deployment
+
+### LMDeploy
+
+LMDeploy is a toolkit for compressing, deploying, and serving LLMs & VLMs.
+
+```sh
+# if lmdeploy<0.7.3, you need to explicitly set chat_template_config=ChatTemplateConfig(model_name='internvl2_5')
+pip install lmdeploy>=0.7.3
+```
+
+LMDeploy abstracts the complex inference process of multi-modal Vision-Language Models (VLM) into an easy-to-use pipeline, similar to the Large Language Model (LLM) inference pipeline.
+
+#### A 'Hello, world' Example
+
+```python
+from lmdeploy import pipeline, TurbomindEngineConfig, ChatTemplateConfig
+from lmdeploy.vl import load_image
+
+model = 'OpenGVLab/InternVL3-14B'
+image = load_image('https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/tests/data/tiger.jpeg')
+pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=16384, tp=1), chat_template_config=ChatTemplateConfig(model_name='internvl2_5'))
+response = pipe(('describe this image', image))
+print(response.text)
+```
+
+If `ImportError` occurs while executing this case, please install the required dependency packages as prompted.
+
+#### Multi-images Inference
+
+When dealing with multiple images, you can put them all in one list. Keep in mind that multiple images will lead to a higher number of input tokens, and as a result, the size of the context window typically needs to be increased.
+
+```python
+from lmdeploy import pipeline, TurbomindEngineConfig, ChatTemplateConfig
+from lmdeploy.vl import load_image
+from lmdeploy.vl.constants import IMAGE_TOKEN
+
+model = 'OpenGVLab/InternVL3-14B'
+pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=16384, tp=1), chat_template_config=ChatTemplateConfig(model_name='internvl2_5'))
+
+image_urls=[
+    'https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/human-pose.jpg',
+    'https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/det.jpg'
+]
+
+images = [load_image(img_url) for img_url in image_urls]
+# Numbering images improves multi-image conversations
+response = pipe((f'Image-1: {IMAGE_TOKEN}\nImage-2: {IMAGE_TOKEN}\ndescribe these two images', images))
+print(response.text)
+```
+
+#### Batch Prompts Inference
+
+Conducting inference with batch prompts is quite straightforward; just place them within a list structure:
+
+```python
+from lmdeploy import pipeline, TurbomindEngineConfig, ChatTemplateConfig
+from lmdeploy.vl import load_image
+
+model = 'OpenGVLab/InternVL3-14B'
+pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=16384, tp=1), chat_template_config=ChatTemplateConfig(model_name='internvl2_5'))
+
+image_urls=[
+    "https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/human-pose.jpg",
+    "https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/det.jpg"
+]
+prompts = [('describe this image', load_image(img_url)) for img_url in image_urls]
+response = pipe(prompts)
+print(response)
+```
+
+#### Multi-turn Conversation
+
+There are two ways to do the multi-turn conversations with the pipeline. One is to construct messages according to the format of OpenAI and use above introduced method, the other is to use the `pipeline.chat` interface.
+
+```python
+from lmdeploy import pipeline, TurbomindEngineConfig, GenerationConfig, ChatTemplateConfig
+from lmdeploy.vl import load_image
+
+model = 'OpenGVLab/InternVL3-14B'
+pipe = pipeline(model, backend_config=TurbomindEngineConfig(session_len=16384, tp=1), chat_template_config=ChatTemplateConfig(model_name='internvl2_5'))
+
+image = load_image('https://raw.githubusercontent.com/open-mmlab/mmdeploy/main/demo/resources/human-pose.jpg')
+gen_config = GenerationConfig(top_k=40, top_p=0.8, temperature=0.8)
+sess = pipe.chat(('describe this image', image), gen_config=gen_config)
+print(sess.response.text)
+sess = pipe.chat('What is the woman doing?', session=sess, gen_config=gen_config)
+print(sess.response.text)
+```
+
+#### Service
+
+LMDeploy's `api_server` enables models to be easily packed into services with a single command. The provided RESTful APIs are compatible with OpenAI's interfaces. Below are an example of service startup:
+
+```shell
+lmdeploy serve api_server OpenGVLab/InternVL3-14B --chat-template internvl2_5 --server-port 23333 --tp 1
+```
+
+To use the OpenAI-style interface, you need to install OpenAI:
+
+```shell
+pip install openai
+```
+
+Then, use the code below to make the API call:
+
+```python
+from openai import OpenAI
+
+client = OpenAI(api_key='YOUR_API_KEY', base_url='http://0.0.0.0:23333/v1')
+model_name = client.models.list().data[0].id
+response = client.chat.completions.create(
+    model=model_name,
+    messages=[{
+        'role':
+        'user',
+        'content': [{
+            'type': 'text',
+            'text': 'describe this image',
+        }, {
+            'type': 'image_url',
+            'image_url': {
+                'url':
+                'https://modelscope.oss-cn-beijing.aliyuncs.com/resource/tiger.jpeg',
+            },
+        }],
+    }],
+    temperature=0.8,
+    top_p=0.8)
+print(response)
+```
+
+## License
+
+This project is released under the MIT License. This project uses the pre-trained Qwen2.5 as a component, which is licensed under the Apache-2.0 License.
+
+## Citation
+
+If you find this project useful in your research, please consider citing:
+
+```BibTeX
+@article{chen2024expanding,
+  title={Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling},
+  author={Chen, Zhe and Wang, Weiyun and Cao, Yue and Liu, Yangzhou and Gao, Zhangwei and Cui, Erfei and Zhu, Jinguo and Ye, Shenglong and Tian, Hao and Liu, Zhaoyang and others},
+  journal={arXiv preprint arXiv:2412.05271},
+  year={2024}
+}
+@article{wang2024mpo,
+  title={Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization},
+  author={Wang, Weiyun and Chen, Zhe and Wang, Wenhai and Cao, Yue and Liu, Yangzhou and Gao, Zhangwei and Zhu, Jinguo and Zhu, Xizhou and Lu, Lewei and Qiao, Yu and Dai, Jifeng},
+  journal={arXiv preprint arXiv:2411.10442},
+  year={2024}
+}
+@article{chen2024far,
+  title={How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites},
+  author={Chen, Zhe and Wang, Weiyun and Tian, Hao and Ye, Shenglong and Gao, Zhangwei and Cui, Erfei and Tong, Wenwen and Hu, Kongzhi and Luo, Jiapeng and Ma, Zheng and others},
+  journal={arXiv preprint arXiv:2404.16821},
+  year={2024}
+}
+@inproceedings{chen2024internvl,
+  title={Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks},
+  author={Chen, Zhe and Wu, Jiannan and Wang, Wenhai and Su, Weijie and Chen, Guo and Xing, Sen and Zhong, Muyan and Zhang, Qinglong and Zhu, Xizhou and Lu, Lewei and others},
+  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
+  pages={24185--24198},
+  year={2024}
+}
+```

added_tokens.json

@@ -0,0 +1,33 @@
+{
+  "</box>": 151673,
+  "</img>": 151666,
+  "</quad>": 151669,
+  "</ref>": 151671,
+  "</tool_call>": 151658,
+  "<IMG_CONTEXT>": 151667,
+  "<box>": 151672,
+  "<img>": 151665,
+  "<quad>": 151668,
+  "<ref>": 151670,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

config.json

@@ -0,0 +1,224 @@
+{
+  "_commit_hash": null,
+  "_name_or_path": "/mnt/petrelfs/wangweiyun/workspace_wwy/open_source/InternVL/internvl_chat/work_dirs/internvl_chat_v3_0/InternVL3_0-14B-MPO-try0-2",
+  "architectures": [
+    "InternVLChatModel"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_internvl_chat.InternVLChatConfig",
+    "AutoModel": "modeling_internvl_chat.InternVLChatModel",
+    "AutoModelForCausalLM": "modeling_internvl_chat.InternVLChatModel"
+  },
+  "downsample_ratio": 0.5,
+  "dynamic_image_size": true,
+  "force_image_size": 448,
+  "hidden_size": 5120,
+  "image_fold": null,
+  "llm_config": {
+    "_attn_implementation_autoset": true,
+    "_name_or_path": "./pretrained/Qwen2.5-32B-Instruct",
+    "add_cross_attention": false,
+    "architectures": [
+      "Qwen2ForCausalLM"
+    ],
+    "attention_dropout": 0.0,
+    "bad_words_ids": null,
+    "begin_suppress_tokens": null,
+    "bos_token_id": 151643,
+    "chunk_size_feed_forward": 0,
+    "cross_attention_hidden_size": null,
+    "decoder_start_token_id": null,
+    "diversity_penalty": 0.0,
+    "do_sample": false,
+    "early_stopping": false,
+    "encoder_no_repeat_ngram_size": 0,
+    "eos_token_id": 151643,
+    "exponential_decay_length_penalty": null,
+    "finetuning_task": null,
+    "forced_bos_token_id": null,
+    "forced_eos_token_id": null,
+    "hidden_act": "silu",
+    "hidden_size": 5120,
+    "id2label": {
+      "0": "LABEL_0",
+      "1": "LABEL_1"
+    },
+    "initializer_range": 0.02,
+    "intermediate_size": 13824,
+    "is_decoder": false,
+    "is_encoder_decoder": false,
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1
+    },
+    "length_penalty": 1.0,
+    "max_length": 20,
+    "max_position_embeddings": 32768,
+    "max_window_layers": 70,
+    "min_length": 0,
+    "model_type": "qwen2",
+    "moe_config": null,
+    "no_repeat_ngram_size": 0,
+    "num_attention_heads": 40,
+    "num_beam_groups": 1,
+    "num_beams": 1,
+    "num_hidden_layers": 48,
+    "num_key_value_heads": 8,
+    "num_return_sequences": 1,
+    "output_attentions": false,
+    "output_hidden_states": false,
+    "output_scores": false,
+    "pad_token_id": null,
+    "prefix": null,
+    "problem_type": null,
+    "pruned_heads": {},
+    "remove_invalid_values": false,
+    "repetition_penalty": 1.0,
+    "return_dict": true,
+    "return_dict_in_generate": false,
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": {
+      "factor": 2.0,
+      "rope_type": "dynamic",
+      "type": "dynamic"
+    },
+    "rope_theta": 1000000.0,
+    "sep_token_id": null,
+    "sliding_window": null,
+    "suppress_tokens": null,
+    "task_specific_params": null,
+    "temperature": 1.0,
+    "tf_legacy_loss": false,
+    "tie_encoder_decoder": false,
+    "tie_word_embeddings": false,
+    "tokenizer_class": null,
+    "top_k": 50,
+    "top_p": 1.0,
+    "torch_dtype": "bfloat16",
+    "torchscript": false,
+    "transformers_version": "4.48.3",
+    "typical_p": 1.0,
+    "use_bfloat16": true,
+    "use_cache": false,
+    "use_sliding_window": false,
+    "vocab_size": 151674
+  },
+  "max_dynamic_patch": 12,
+  "min_dynamic_patch": 1,
+  "model_type": "internvl_chat",
+  "pad2square": false,
+  "ps_version": "v2",
+  "select_layer": -1,
+  "system_message": null,
+  "template": "internvl2_5",
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": null,
+  "use_backbone_lora": 0,
+  "use_llm_lora": 0,
+  "use_thumbnail": true,
+  "vision_config": {
+    "_attn_implementation_autoset": true,
+    "_name_or_path": "OpenGVLab/InternViT-6B-448px-V1-5",
+    "add_cross_attention": false,
+    "architectures": [
+      "InternVisionModel"
+    ],
+    "attention_dropout": 0.0,
+    "auto_map": {
+      "AutoConfig": "configuration_intern_vit.InternVisionConfig",
+      "AutoModel": "modeling_intern_vit.InternVisionModel"
+    },
+    "bad_words_ids": null,
+    "begin_suppress_tokens": null,
+    "bos_token_id": null,
+    "capacity_factor": 1.2,
+    "chunk_size_feed_forward": 0,
+    "cross_attention_hidden_size": null,
+    "decoder_start_token_id": null,
+    "diversity_penalty": 0.0,
+    "do_sample": false,
+    "drop_path_rate": 0.1,
+    "dropout": 0.0,
+    "early_stopping": false,
+    "encoder_no_repeat_ngram_size": 0,
+    "eos_token_id": null,
+    "eval_capacity_factor": 1.4,
+    "exponential_decay_length_penalty": null,
+    "finetuning_task": null,
+    "forced_bos_token_id": null,
+    "forced_eos_token_id": null,
+    "hidden_act": "gelu",
+    "hidden_size": 1024,
+    "id2label": {
+      "0": "LABEL_0",
+      "1": "LABEL_1"
+    },
+    "image_size": 448,
+    "initializer_factor": 0.1,
+    "initializer_range": 1e-10,
+    "intermediate_size": 4096,
+    "is_decoder": false,
+    "is_encoder_decoder": false,
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1
+    },
+    "laux_allreduce": "all_nodes",
+    "layer_norm_eps": 1e-06,
+    "length_penalty": 1.0,
+    "max_length": 20,
+    "min_length": 0,
+    "model_type": "intern_vit_6b",
+    "moe_coeff_ratio": 0.5,
+    "moe_intermediate_size": 768,
+    "moe_output_scale": 4.0,
+    "no_repeat_ngram_size": 0,
+    "noisy_gate_policy": "RSample_before",
+    "norm_type": "layer_norm",
+    "num_attention_heads": 16,
+    "num_beam_groups": 1,
+    "num_beams": 1,
+    "num_channels": 3,
+    "num_experts": 8,
+    "num_hidden_layers": 24,
+    "num_return_sequences": 1,
+    "num_routed_experts": 4,
+    "num_shared_experts": 4,
+    "output_attentions": false,
+    "output_hidden_states": false,
+    "output_scores": false,
+    "pad_token_id": null,
+    "patch_size": 14,
+    "prefix": null,
+    "problem_type": null,
+    "pruned_heads": {},
+    "qk_normalization": false,
+    "qkv_bias": true,
+    "remove_invalid_values": false,
+    "repetition_penalty": 1.0,
+    "return_dict": true,
+    "return_dict_in_generate": false,
+    "sep_token_id": null,
+    "shared_expert_intermediate_size": 3072,
+    "suppress_tokens": null,
+    "task_specific_params": null,
+    "temperature": 1.0,
+    "tf_legacy_loss": false,
+    "tie_encoder_decoder": false,
+    "tie_word_embeddings": true,
+    "tokenizer_class": null,
+    "top_k": 50,
+    "top_p": 1.0,
+    "torch_dtype": "bfloat16",
+    "torchscript": false,
+    "transformers_version": "4.48.3",
+    "typical_p": 1.0,
+    "use_bfloat16": true,
+    "use_flash_attn": true,
+    "use_moe": false,
+    "use_residual": true,
+    "use_rts": false,
+    "use_weighted_residual": false
+  }
+}

configuration_intern_vit.py

@@ -0,0 +1,120 @@
+# --------------------------------------------------------
+# InternVL
+# Copyright (c) 2024 OpenGVLab
+# Licensed under The MIT License [see LICENSE for details]
+# --------------------------------------------------------
+
+import os
+from typing import Union
+
+from transformers.configuration_utils import PretrainedConfig
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+
+class InternVisionConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`InternVisionModel`]. It is used to
+    instantiate a vision encoder according to the specified arguments, defining the model architecture.
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+    Args:
+        num_channels (`int`, *optional*, defaults to 3):
+            Number of color channels in the input images (e.g., 3 for RGB).
+        patch_size (`int`, *optional*, defaults to 14):
+            The size (resolution) of each patch.
+        image_size (`int`, *optional*, defaults to 224):
+            The size (resolution) of each image.
+        qkv_bias (`bool`, *optional*, defaults to `False`):
+            Whether to add a bias to the queries and values in the self-attention layers.
+        hidden_size (`int`, *optional*, defaults to 3200):
+            Dimensionality of the encoder layers and the pooler layer.
+        num_attention_heads (`int`, *optional*, defaults to 25):
+            Number of attention heads for each attention layer in the Transformer encoder.
+        intermediate_size (`int`, *optional*, defaults to 12800):
+            Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder.
+        qk_normalization (`bool`, *optional*, defaults to `True`):
+            Whether to normalize the queries and keys in the self-attention layers.
+        num_hidden_layers (`int`, *optional*, defaults to 48):
+            Number of hidden layers in the Transformer encoder.
+        use_flash_attn (`bool`, *optional*, defaults to `True`):
+            Whether to use flash attention mechanism.
+        hidden_act (`str` or `function`, *optional*, defaults to `"gelu"`):
+            The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
+            `"relu"`, `"selu"` and `"gelu_new"` ``"gelu"` are supported.
+        layer_norm_eps (`float`, *optional*, defaults to 1e-6):
+            The epsilon used by the layer normalization layers.
+        dropout (`float`, *optional*, defaults to 0.0):
+            The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
+        drop_path_rate (`float`, *optional*, defaults to 0.0):
+            Dropout rate for stochastic depth.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        initializer_factor (`float`, *optional*, defaults to 0.1):
+            A factor for layer scale.
+    """
+
+    model_type = 'intern_vit_6b'
+
+    def __init__(
+            self,
+            num_channels=3,
+            patch_size=14,
+            image_size=224,
+            qkv_bias=False,
+            hidden_size=3200,
+            num_attention_heads=25,
+            intermediate_size=12800,
+            qk_normalization=True,
+            num_hidden_layers=48,
+            use_flash_attn=True,
+            hidden_act='gelu',
+            norm_type='rms_norm',
+            layer_norm_eps=1e-6,
+            dropout=0.0,
+            drop_path_rate=0.0,
+            attention_dropout=0.0,
+            initializer_range=0.02,
+            initializer_factor=0.1,
+            **kwargs,
+    ):
+        super().__init__(**kwargs)
+
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.dropout = dropout
+        self.drop_path_rate = drop_path_rate
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.num_channels = num_channels
+        self.patch_size = patch_size
+        self.image_size = image_size
+        self.initializer_range = initializer_range
+        self.initializer_factor = initializer_factor
+        self.attention_dropout = attention_dropout
+        self.layer_norm_eps = layer_norm_eps
+        self.hidden_act = hidden_act
+        self.norm_type = norm_type
+        self.qkv_bias = qkv_bias
+        self.qk_normalization = qk_normalization
+        self.use_flash_attn = use_flash_attn
+
+    @classmethod
+    def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os.PathLike], **kwargs) -> 'PretrainedConfig':
+        config_dict, kwargs = cls.get_config_dict(pretrained_model_name_or_path, **kwargs)
+
+        if 'vision_config' in config_dict:
+            config_dict = config_dict['vision_config']
+
+        if 'model_type' in config_dict and hasattr(cls, 'model_type') and config_dict['model_type'] != cls.model_type:
+            logger.warning(
+                f"You are using a model of type {config_dict['model_type']} to instantiate a model of type "
+                f'{cls.model_type}. This is not supported for all configurations of models and can yield errors.'
+            )
+
+        return cls.from_dict(config_dict, **kwargs)

configuration_internvl_chat.py

@@ -0,0 +1,97 @@
+# --------------------------------------------------------
+# InternVL
+# Copyright (c) 2024 OpenGVLab
+# Licensed under The MIT License [see LICENSE for details]
+# --------------------------------------------------------
+
+import copy
+
+from transformers import AutoConfig, LlamaConfig, Qwen2Config
+from transformers.configuration_utils import PretrainedConfig
+from transformers.utils import logging
+
+from .configuration_intern_vit import InternVisionConfig
+
+logger = logging.get_logger(__name__)
+
+
+class InternVLChatConfig(PretrainedConfig):
+    model_type = 'internvl_chat'
+    is_composition = True
+
+    def __init__(
+            self,
+            vision_config=None,
+            llm_config=None,
+            use_backbone_lora=0,
+            use_llm_lora=0,
+            select_layer=-1,
+            force_image_size=None,
+            downsample_ratio=0.5,
+            template=None,
+            dynamic_image_size=False,
+            use_thumbnail=False,
+            ps_version='v1',
+            min_dynamic_patch=1,
+            max_dynamic_patch=6,
+            **kwargs):
+        super().__init__(**kwargs)
+
+        if vision_config is None:
+            vision_config = {'architectures': ['InternVisionModel']}
+            logger.info('vision_config is None. Initializing the InternVisionConfig with default values.')
+
+        if llm_config is None:
+            llm_config = {'architectures': ['Qwen2ForCausalLM']}
+            logger.info('llm_config is None. Initializing the LlamaConfig config with default values (`LlamaConfig`).')
+
+        self.vision_config = InternVisionConfig(**vision_config)
+        if llm_config.get('architectures')[0] == 'LlamaForCausalLM':
+            self.llm_config = LlamaConfig(**llm_config)
+        elif llm_config.get('architectures')[0] == 'Qwen2ForCausalLM':
+            self.llm_config = Qwen2Config(**llm_config)
+        else:
+            raise ValueError('Unsupported architecture: {}'.format(llm_config.get('architectures')[0]))
+        self.use_backbone_lora = use_backbone_lora
+        self.use_llm_lora = use_llm_lora
+        self.select_layer = select_layer
+        self.force_image_size = force_image_size
+        self.downsample_ratio = downsample_ratio
+        self.template = template
+        self.dynamic_image_size = dynamic_image_size
+        self.use_thumbnail = use_thumbnail
+        self.ps_version = ps_version  # pixel shuffle version
+        self.min_dynamic_patch = min_dynamic_patch
+        self.max_dynamic_patch = max_dynamic_patch
+        # By default, we use tie_word_embeddings=False for models of all sizes.
+        self.tie_word_embeddings = self.llm_config.tie_word_embeddings
+
+        logger.info(f'vision_select_layer: {self.select_layer}')
+        logger.info(f'ps_version: {self.ps_version}')
+        logger.info(f'min_dynamic_patch: {self.min_dynamic_patch}')
+        logger.info(f'max_dynamic_patch: {self.max_dynamic_patch}')
+
+    def to_dict(self):
+        """
+        Serializes this instance to a Python dictionary. Override the default [`~PretrainedConfig.to_dict`].
+
+        Returns:
+            `Dict[str, any]`: Dictionary of all the attributes that make up this configuration instance,
+        """
+        output = copy.deepcopy(self.__dict__)
+        output['vision_config'] = self.vision_config.to_dict()
+        output['llm_config'] = self.llm_config.to_dict()
+        output['model_type'] = self.__class__.model_type
+        output['use_backbone_lora'] = self.use_backbone_lora
+        output['use_llm_lora'] = self.use_llm_lora
+        output['select_layer'] = self.select_layer
+        output['force_image_size'] = self.force_image_size
+        output['downsample_ratio'] = self.downsample_ratio
+        output['template'] = self.template
+        output['dynamic_image_size'] = self.dynamic_image_size
+        output['use_thumbnail'] = self.use_thumbnail
+        output['ps_version'] = self.ps_version
+        output['min_dynamic_patch'] = self.min_dynamic_patch
+        output['max_dynamic_patch'] = self.max_dynamic_patch
+
+        return output

conversation.py

@@ -0,0 +1,391 @@
+"""
+Conversation prompt templates.
+
+We kindly request that you import fastchat instead of copying this file if you wish to use it.
+If you have changes in mind, please contribute back so the community can benefit collectively and continue to maintain these valuable templates.
+
+Modified from https://github.com/lm-sys/FastChat/blob/main/fastchat/conversation.py
+"""
+
+import dataclasses
+from enum import IntEnum, auto
+from typing import Dict, List, Tuple, Union
+
+
+class SeparatorStyle(IntEnum):
+    """Separator styles."""
+
+    ADD_COLON_SINGLE = auto()
+    ADD_COLON_TWO = auto()
+    ADD_COLON_SPACE_SINGLE = auto()
+    NO_COLON_SINGLE = auto()
+    NO_COLON_TWO = auto()
+    ADD_NEW_LINE_SINGLE = auto()
+    LLAMA2 = auto()
+    CHATGLM = auto()
+    CHATML = auto()
+    CHATINTERN = auto()
+    DOLLY = auto()
+    RWKV = auto()
+    PHOENIX = auto()
+    ROBIN = auto()
+    FALCON_CHAT = auto()
+    CHATGLM3 = auto()
+    INTERNVL_ZH = auto()
+    MPT = auto()
+
+
+@dataclasses.dataclass
+class Conversation:
+    """A class that manages prompt templates and keeps all conversation history."""
+
+    # The name of this template
+    name: str
+    # The template of the system prompt
+    system_template: str = '{system_message}'
+    # The system message
+    system_message: str = ''
+    # The names of two roles
+    roles: Tuple[str] = ('USER', 'ASSISTANT')
+    # All messages. Each item is (role, message).
+    messages: List[List[str]] = ()
+    # The number of few shot examples
+    offset: int = 0
+    # The separator style and configurations
+    sep_style: SeparatorStyle = SeparatorStyle.ADD_COLON_SINGLE
+    sep: str = '\n'
+    sep2: str = None
+    # Stop criteria (the default one is EOS token)
+    stop_str: Union[str, List[str]] = None
+    # Stops generation if meeting any token in this list
+    stop_token_ids: List[int] = None
+
+    def get_prompt(self) -> str:
+        """Get the prompt for generation."""
+        system_prompt = self.system_template.format(system_message=self.system_message)
+        if self.sep_style == SeparatorStyle.ADD_COLON_SINGLE:
+            ret = system_prompt + self.sep
+            for role, message in self.messages:
+                if message:
+                    ret += role + ': ' + message + self.sep
+                else:
+                    ret += role + ':'
+            return ret
+        elif self.sep_style == SeparatorStyle.ADD_COLON_TWO:
+            seps = [self.sep, self.sep2]
+            ret = system_prompt + seps[0]
+            for i, (role, message) in enumerate(self.messages):
+                if message:
+                    ret += role + ': ' + message + seps[i % 2]
+                else:
+                    ret += role + ':'
+            return ret
+        elif self.sep_style == SeparatorStyle.ADD_COLON_SPACE_SINGLE:
+            ret = system_prompt + self.sep
+            for role, message in self.messages:
+                if message:
+                    ret += role + ': ' + message + self.sep
+                else:
+                    ret += role + ': '  # must be end with a space
+            return ret
+        elif self.sep_style == SeparatorStyle.ADD_NEW_LINE_SINGLE:
+            ret = '' if system_prompt == '' else system_prompt + self.sep
+            for role, message in self.messages:
+                if message:
+                    ret += role + '\n' + message + self.sep
+                else:
+                    ret += role + '\n'
+            return ret
+        elif self.sep_style == SeparatorStyle.NO_COLON_SINGLE:
+            ret = system_prompt
+            for role, message in self.messages:
+                if message:
+                    ret += role + message + self.sep
+                else:
+                    ret += role
+            return ret
+        elif self.sep_style == SeparatorStyle.NO_COLON_TWO:
+            seps = [self.sep, self.sep2]
+            ret = system_prompt
+            for i, (role, message) in enumerate(self.messages):
+                if message:
+                    ret += role + message + seps[i % 2]
+                else:
+                    ret += role
+            return ret
+        elif self.sep_style == SeparatorStyle.RWKV:
+            ret = system_prompt
+            for i, (role, message) in enumerate(self.messages):
+                if message:
+                    ret += (
+                        role
+                        + ': '
+                        + message.replace('\r\n', '\n').replace('\n\n', '\n')
+                    )
+                    ret += '\n\n'
+                else:
+                    ret += role + ':'
+            return ret
+        elif self.sep_style == SeparatorStyle.LLAMA2:
+            seps = [self.sep, self.sep2]
+            if self.system_message:
+                ret = system_prompt
+            else:
+                ret = '[INST] '
+            for i, (role, message) in enumerate(self.messages):
+                tag = self.roles[i % 2]
+                if message:
+                    if i == 0:
+                        ret += message + ' '
+                    else:
+                        ret += tag + ' ' + message + seps[i % 2]
+                else:
+                    ret += tag
+            return ret
+        elif self.sep_style == SeparatorStyle.CHATGLM:
+            # source: https://huggingface.co/THUDM/chatglm-6b/blob/1d240ba371910e9282298d4592532d7f0f3e9f3e/modeling_chatglm.py#L1302-L1308
+            # source2: https://huggingface.co/THUDM/chatglm2-6b/blob/e186c891cf64310ac66ef10a87e6635fa6c2a579/modeling_chatglm.py#L926
+            round_add_n = 1 if self.name == 'chatglm2' else 0
+            if system_prompt:
+                ret = system_prompt + self.sep
+            else:
+                ret = ''
+
+            for i, (role, message) in enumerate(self.messages):
+                if i % 2 == 0:
+                    ret += f'[Round {i//2 + round_add_n}]{self.sep}'
+
+                if message:
+                    ret += f'{role}：{message}{self.sep}'
+                else:
+                    ret += f'{role}：'
+            return ret
+        elif self.sep_style == SeparatorStyle.CHATML:
+            ret = '' if system_prompt == '' else system_prompt + self.sep + '\n'
+            for role, message in self.messages:
+                if message:
+                    ret += role + '\n' + message + self.sep + '\n'
+                else:
+                    ret += role + '\n'
+            return ret
+        elif self.sep_style == SeparatorStyle.CHATGLM3:
+            ret = ''
+            if self.system_message:
+                ret += system_prompt
+            for role, message in self.messages:
+                if message:
+                    ret += role + '\n' + ' ' + message
+                else:
+                    ret += role
+            return ret
+        elif self.sep_style == SeparatorStyle.CHATINTERN:
+            # source: https://huggingface.co/internlm/internlm-chat-7b-8k/blob/bd546fa984b4b0b86958f56bf37f94aa75ab8831/modeling_internlm.py#L771
+            seps = [self.sep, self.sep2]
+            ret = system_prompt
+            for i, (role, message) in enumerate(self.messages):
+                # if i % 2 == 0:
+                #     ret += "<s>"
+                if message:
+                    ret += role + ':' + message + seps[i % 2] + '\n'
+                else:
+                    ret += role + ':'
+            return ret
+        elif self.sep_style == SeparatorStyle.DOLLY:
+            seps = [self.sep, self.sep2]
+            ret = system_prompt
+            for i, (role, message) in enumerate(self.messages):
+                if message:
+                    ret += role + ':\n' + message + seps[i % 2]
+                    if i % 2 == 1:
+                        ret += '\n\n'
+                else:
+                    ret += role + ':\n'
+            return ret
+        elif self.sep_style == SeparatorStyle.PHOENIX:
+            ret = system_prompt
+            for role, message in self.messages:
+                if message:
+                    ret += role + ': ' + '<s>' + message + '</s>'
+                else:
+                    ret += role + ': ' + '<s>'
+            return ret
+        elif self.sep_style == SeparatorStyle.ROBIN:
+            ret = system_prompt + self.sep
+            for role, message in self.messages:
+                if message:
+                    ret += role + ':\n' + message + self.sep
+                else:
+                    ret += role + ':\n'
+            return ret
+        elif self.sep_style == SeparatorStyle.FALCON_CHAT:
+            ret = ''
+            if self.system_message:
+                ret += system_prompt + self.sep
+            for role, message in self.messages:
+                if message:
+                    ret += role + ': ' + message + self.sep
+                else:
+                    ret += role + ':'
+
+            return ret
+        elif self.sep_style == SeparatorStyle.INTERNVL_ZH:
+            seps = [self.sep, self.sep2]
+            ret = self.system_message + seps[0]
+            for i, (role, message) in enumerate(self.messages):
+                if message:
+                    ret += role + ': ' + message + seps[i % 2]
+                else:
+                    ret += role + ':'
+            return ret
+        elif self.sep_style == SeparatorStyle.MPT:
+            ret = system_prompt + self.sep
+            for role, message in self.messages:
+                if message:
+                    if type(message) is tuple:
+                        message, _, _ = message
+                    ret += role + message + self.sep
+                else:
+                    ret += role
+            return ret
+        else:
+            raise ValueError(f'Invalid style: {self.sep_style}')
+
+    def set_system_message(self, system_message: str):
+        """Set the system message."""
+        self.system_message = system_message
+
+    def append_message(self, role: str, message: str):
+        """Append a new message."""
+        self.messages.append([role, message])
+
+    def update_last_message(self, message: str):
+        """Update the last output.
+
+        The last message is typically set to be None when constructing the prompt,
+        so we need to update it in-place after getting the response from a model.
+        """
+        self.messages[-1][1] = message
+
+    def to_gradio_chatbot(self):
+        """Convert the conversation to gradio chatbot format."""
+        ret = []
+        for i, (role, msg) in enumerate(self.messages[self.offset :]):
+            if i % 2 == 0:
+                ret.append([msg, None])
+            else:
+                ret[-1][-1] = msg
+        return ret
+
+    def to_openai_api_messages(self):
+        """Convert the conversation to OpenAI chat completion format."""
+        ret = [{'role': 'system', 'content': self.system_message}]
+
+        for i, (_, msg) in enumerate(self.messages[self.offset :]):
+            if i % 2 == 0:
+                ret.append({'role': 'user', 'content': msg})
+            else:
+                if msg is not None:
+                    ret.append({'role': 'assistant', 'content': msg})
+        return ret
+
+    def copy(self):
+        return Conversation(
+            name=self.name,
+            system_template=self.system_template,
+            system_message=self.system_message,
+            roles=self.roles,
+            messages=[[x, y] for x, y in self.messages],
+            offset=self.offset,
+            sep_style=self.sep_style,
+            sep=self.sep,
+            sep2=self.sep2,
+            stop_str=self.stop_str,
+            stop_token_ids=self.stop_token_ids,
+        )
+
+    def dict(self):
+        return {
+            'template_name': self.name,
+            'system_message': self.system_message,
+            'roles': self.roles,
+            'messages': self.messages,
+            'offset': self.offset,
+        }
+
+
+# A global registry for all conversation templates
+conv_templates: Dict[str, Conversation] = {}
+
+
+def register_conv_template(template: Conversation, override: bool = False):
+    """Register a new conversation template."""
+    if not override:
+        assert (
+            template.name not in conv_templates
+        ), f'{template.name} has been registered.'
+
+    conv_templates[template.name] = template
+
+
+def get_conv_template(name: str) -> Conversation:
+    """Get a conversation template."""
+    return conv_templates[name].copy()
+
+
+# Both Hermes-2 and internlm2-chat are chatml-format conversation templates. The difference
+# is that during training, the preprocessing function for the Hermes-2 template doesn't add
+# <s> at the beginning of the tokenized sequence, while the internlm2-chat template does.
+# Therefore, they are completely equivalent during inference.
+register_conv_template(
+    Conversation(
+        name='Hermes-2',
+        system_template='<|im_start|>system\n{system_message}',
+        # note: The new system prompt was not used here to avoid changes in benchmark performance.
+        # system_message='我是书生·万象，英文名是InternVL，是由上海人工智能实验室、清华大学及多家合作单位联合开发的多模态大语言模型。',
+        system_message='你是由上海人工智能实验室联合商汤科技开发的书生多模态大模型，英文名叫InternVL, 是一个有用无害的人工智能助手。',
+        roles=('<|im_start|>user\n', '<|im_start|>assistant\n'),
+        sep_style=SeparatorStyle.MPT,
+        sep='<|im_end|>',
+        stop_str='<|endoftext|>',
+    )
+)
+
+
+register_conv_template(
+    Conversation(
+        name='internlm2-chat',
+        system_template='<|im_start|>system\n{system_message}',
+        # note: The new system prompt was not used here to avoid changes in benchmark performance.
+        # system_message='我是书生·万象，英文名是InternVL，是由上海人工智能实验室、清华大学及多家合作单位联合开发的多模态大语言模型。',
+        system_message='你是由上海人工智能实验室联合商汤科技开发的书生多模态大模型，英文名叫InternVL, 是一个有用无害的人工智能助手。',
+        roles=('<|im_start|>user\n', '<|im_start|>assistant\n'),
+        sep_style=SeparatorStyle.MPT,
+        sep='<|im_end|>',
+    )
+)
+
+
+register_conv_template(
+    Conversation(
+        name='phi3-chat',
+        system_template='<|system|>\n{system_message}',
+        # note: The new system prompt was not used here to avoid changes in benchmark performance.
+        # system_message='我是书生·万象，英文名是InternVL，是由上海人工智能实验室、清华大学及多家合作单位联合开发的多模态大语言模型。',
+        system_message='你是由上海人工智能实验室联合商汤科技开发的书生多模态大模型，英文名叫InternVL, 是一个有用无害的人工智能助手。',
+        roles=('<|user|>\n', '<|assistant|>\n'),
+        sep_style=SeparatorStyle.MPT,
+        sep='<|end|>',
+    )
+)
+
+
+register_conv_template(
+    Conversation(
+        name='internvl2_5',
+        system_template='<|im_start|>system\n{system_message}',
+        system_message='你是书生·万象，英文名是InternVL，是由上海人工智能实验室、清华大学及多家合作单位联合开发的多模态大语言模型。',
+        roles=('<|im_start|>user\n', '<|im_start|>assistant\n'),
+        sep_style=SeparatorStyle.MPT,
+        sep='<|im_end|>\n',
+    )
+)

model-00001-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af141caa3e661d782379fe7ed38de89cdb551ba731ff38431065904138442d00
+size 5308449216

model-00002-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be62caa795c3b56d70ce2a0288ded43756be634c911977b7e7a38ef1be8d7f18
+size 5237964760

model-00003-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f180048b33eb7f2e65a51c2c04413ed6f3c097736bce6389fbb9111187e2d464
+size 5363829864

model-00004-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a416ea8ee8c4eaf77761eba084cf59811d7ef2b960169aa21c08442c0693de44
+size 5363829856

model-00005-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fdea041ed996827f952692a1db83426f8016e58ec24741fea06a8640843b5c9
+size 5237964760

model-00006-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c15947acf6d3e2e2dd776b625fb4d4ea88bf38f475257fc48b492b47c09a14a
+size 3722590528

model.safetensors.index.json

@@ -0,0 +1,933 @@
+{
+  "metadata": {
+    "total_size": 30234513408,
+    "format": "pt"
+  },
+  "weight_map": {
+    "language_model.lm_head.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.embed_tokens.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.self_attn.k_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.self_attn.q_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.self_attn.v_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.self_attn.k_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.self_attn.q_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.self_attn.v_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.self_attn.k_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.self_attn.q_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.self_attn.v_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.10.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.self_attn.k_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.self_attn.q_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.self_attn.v_proj.bias": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.11.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.12.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "language_model.model.layers.12.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.12.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.12.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.12.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.12.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.12.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.12.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.12.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.12.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.12.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.12.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.13.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.14.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.15.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.16.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.17.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.18.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.19.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.self_attn.k_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.self_attn.q_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.self_attn.v_proj.bias": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.2.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.20.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.20.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.20.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "language_model.model.layers.20.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.20.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.20.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.20.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.20.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.20.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.20.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.20.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.20.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.21.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.22.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.23.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.24.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.25.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.26.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.27.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.28.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.self_attn.k_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.self_attn.q_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.self_attn.v_proj.bias": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.29.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.3.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.3.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "language_model.model.layers.3.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.3.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.3.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.3.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.3.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.3.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.3.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.3.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.3.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.3.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.30.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.31.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.32.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.33.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.34.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.35.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.36.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.37.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.self_attn.k_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.self_attn.q_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.self_attn.v_proj.bias": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.38.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.39.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "language_model.model.layers.39.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.39.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.39.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.39.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.39.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.39.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.39.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.39.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.39.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.39.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.39.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.4.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.40.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.41.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.42.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.43.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.44.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.45.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.self_attn.k_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.self_attn.q_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.self_attn.v_proj.bias": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.46.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.47.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.47.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.47.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "language_model.model.layers.47.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.47.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.47.self_attn.k_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.47.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.47.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.47.self_attn.q_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.47.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.47.self_attn.v_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.47.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.self_attn.k_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.self_attn.q_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.self_attn.v_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.5.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.self_attn.k_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.self_attn.q_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.self_attn.v_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.6.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.self_attn.k_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.self_attn.q_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.self_attn.v_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.7.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.self_attn.k_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.self_attn.q_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.self_attn.v_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.8.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.self_attn.k_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.self_attn.q_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.self_attn.v_proj.bias": "model-00006-of-00006.safetensors",
+    "language_model.model.layers.9.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
+    "language_model.model.norm.weight": "model-00006-of-00006.safetensors",
+    "mlp1.0.bias": "model-00006-of-00006.safetensors",
+    "mlp1.0.weight": "model-00006-of-00006.safetensors",
+    "mlp1.1.bias": "model-00006-of-00006.safetensors",
+    "mlp1.1.weight": "model-00006-of-00006.safetensors",
+    "mlp1.3.bias": "model-00006-of-00006.safetensors",
+    "mlp1.3.weight": "model-00006-of-00006.safetensors",
+    "vision_model.embeddings.class_embedding": "model-00006-of-00006.safetensors",
+    "vision_model.embeddings.patch_embedding.bias": "model-00006-of-00006.safetensors",
+    "vision_model.embeddings.patch_embedding.weight": "model-00006-of-00006.safetensors",
+    "vision_model.embeddings.position_embedding": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.0.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.1.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.10.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.11.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.12.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.13.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.14.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.15.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.16.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.17.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.18.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.19.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.2.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.20.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.21.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.22.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.23.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.3.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.4.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.5.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.6.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.7.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.8.norm2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.attn.proj.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.attn.proj.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.attn.qkv.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.attn.qkv.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.ls1": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.ls2": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.mlp.fc1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.mlp.fc1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.mlp.fc2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.mlp.fc2.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.norm1.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.norm1.weight": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.norm2.bias": "model-00006-of-00006.safetensors",
+    "vision_model.encoder.layers.9.norm2.weight": "model-00006-of-00006.safetensors"
+  }
+}