OpenGVLab
/

InternVL-Chat-V1-1

Image-Text-to-Text

feature-extraction

Model card Files Files and versions Community

czczup commited on Jul 24

Commit

f53243d

•

1 Parent(s): 550a865

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -18,7 +18,7 @@ pipeline_tag: visual-question-answering
 We released InternVL-Chat-V1-1, featuring a structure similar to LLaVA, including a ViT, an MLP projector, and an LLM. As shown in the figure below, we connected our InternViT-6B to LLaMA2-13B through a simple MLP projector. Note that the LLaMA2-13B used here is not the original model but an internal chat version obtained by incrementally pre-training and fine-tuning the LLaMA2-13B base model for Chinese language tasks. Overall, our model has a total of 19 billion parameters.
-<img src="https://cdn-uploads.huggingface.co/production/uploads/64119264f0f81eb569e0d569/HD29tU-g0An9FpQn1yK8X.png" style="width: 80%;">
 In this version, we explored increasing the resolution to 448 × 448, enhancing OCR capabilities, and improving support for Chinese conversations. Since the 448 × 448 input image generates 1024 visual tokens after passing through the ViT, leading to a significant computational burden, we use a pixel shuffle operation to reduce the 1024 tokens to 256 tokens.

 We released InternVL-Chat-V1-1, featuring a structure similar to LLaVA, including a ViT, an MLP projector, and an LLM. As shown in the figure below, we connected our InternViT-6B to LLaMA2-13B through a simple MLP projector. Note that the LLaMA2-13B used here is not the original model but an internal chat version obtained by incrementally pre-training and fine-tuning the LLaMA2-13B base model for Chinese language tasks. Overall, our model has a total of 19 billion parameters.
+<img src="https://cdn-uploads.huggingface.co/production/uploads/64119264f0f81eb569e0d569/HD29tU-g0An9FpQn1yK8X.png" style="width: 100%;">
 In this version, we explored increasing the resolution to 448 × 448, enhancing OCR capabilities, and improving support for Chinese conversations. Since the 448 × 448 input image generates 1024 visual tokens after passing through the ViT, leading to a significant computational burden, we use a pixel shuffle operation to reduce the 1024 tokens to 256 tokens.