Good work. Can you share the following details regarding the pretraining of Supra-50M base model?
- GPU(s) used for pretraining
- Total GPU hours and cost
- Cloud platform (GPU) used for pretraining
Kalyan KS PRO
kalyan-ks
AI & ML interests
NLP (LLMs)
Recent Activity
liked a model about 6 hours ago
principled-intelligence/scope-guard-4B-q-2601 liked a Space 1 day ago
AxiomicLabs/Open_SLM_LeaderboardOrganizations
replied to AxionLab-official's post 1 day ago
reacted to AxionLab-official's post with π 1 day ago
Post
6444
We're happy to announce that we released a Reasoning tuned version of Supra-50M!
SupraLabs/Supra-50M-Reasoning
SupraLabs/Supra-50M-Reasoning
reacted to danielhanchen's post with π 3 days ago
Post
8598
Gemma 4 12B can now run locally on just 8GB RAM via Dynamic GGUFs.
Google's new model, Gemma 4 12B Unified supports image, audio and 256K context.
You can run and train the model via Unsloth Studio.
GGUF: unsloth/gemma-4-12b-it-GGUF
Guide: https://unsloth.ai/docs/models/gemma-4
Google's new model, Gemma 4 12B Unified supports image, audio and 256K context.
You can run and train the model via Unsloth Studio.
GGUF: unsloth/gemma-4-12b-it-GGUF
Guide: https://unsloth.ai/docs/models/gemma-4
- 3 replies
posted an update 12 days ago
Post
1608
LLM Guardrail Models are Less Robust Against Text Mutation Attacks
Blog post - https://huggingface.co/blog/kalyan-ks/llm-guardrail-models-less-robust
Evaluated the robustness of three LLM guardrail models (GLiGuard, LlamaGuard3 and MiniGuard).
Evaluation is done using 16 text mutation attacks over three datasets (AEGIS 2.0, WildGuard and ExpGuard).
Achieved average Unsafe ASR score of up to 33% and average Safe ASR score of up to 25% against GLiGuard model.
Achieved average Unsafe ASR score of up to 35% and average Safe ASR score of up to 17% against LlamaGuard3-8B model.
Achieved average Unsafe ASR score of up to 45% and average Safe ASR score of up to 15% against MiniGuard v0.1 model.
Blog post - https://huggingface.co/blog/kalyan-ks/llm-guardrail-models-less-robust
Evaluated the robustness of three LLM guardrail models (GLiGuard, LlamaGuard3 and MiniGuard).
Evaluation is done using 16 text mutation attacks over three datasets (AEGIS 2.0, WildGuard and ExpGuard).
Achieved average Unsafe ASR score of up to 33% and average Safe ASR score of up to 25% against GLiGuard model.
Achieved average Unsafe ASR score of up to 35% and average Safe ASR score of up to 17% against LlamaGuard3-8B model.
Achieved average Unsafe ASR score of up to 45% and average Safe ASR score of up to 15% against MiniGuard v0.1 model.
reacted to HannesVonEssen's post with β€οΈ 20 days ago
Post
4895
π£ Add architecture visualization to model card!
π For all creators out there: add a model visualization to your model card to capture your audience's attention!
π±οΈ When clicked, it opens an interactive view with multiple levels of granularity!
1οΈβ£ Paste url at https://hfviewer.com/model-card-embed
2οΈβ£ Paste generated code in your README.md!
3οΈβ£ β¨
π For all creators out there: add a model visualization to your model card to capture your audience's attention!
π±οΈ When clicked, it opens an interactive view with multiple levels of granularity!
1οΈβ£ Paste url at https://hfviewer.com/model-card-embed
2οΈβ£ Paste generated code in your README.md!
3οΈβ£ β¨
reacted to danielhanchen's post with β€οΈ 25 days ago
Post
5822
Weβre excited to announce that Unsloth has joined the PyTorch Ecosystem! π₯π¦₯
Unsloth is an open-source project that makes training & running models more accurate and faster with less compute. Our mission is to make local AI accessible to everyone. Thanks to all of you for making this possible! π
Blog: https://unsloth.ai/blog/pytorch
GitHub: https://github.com/unslothai/unsloth
Unsloth is an open-source project that makes training & running models more accurate and faster with less compute. Our mission is to make local AI accessible to everyone. Thanks to all of you for making this possible! π
Blog: https://unsloth.ai/blog/pytorch
GitHub: https://github.com/unslothai/unsloth
- 2 replies
posted an update 26 days ago
Post
1975
Tiny Specialized Models beat LLMs at PII Entity Extraction
Ettin-68M-Nemotron-PII model
- achieves strong F1-score of 96.27% with just 68M parameters.
- beats popular LLMs like GPT-4o-mini (78.69%) and DeepSeek-v4-flash (84.89%).
More details in the blog post - https://huggingface.co/blog/kalyan-ks/tiny-pii-entity-detection-models
Ettin-68M-Nemotron-PII model
- achieves strong F1-score of 96.27% with just 68M parameters.
- beats popular LLMs like GPT-4o-mini (78.69%) and DeepSeek-v4-flash (84.89%).
More details in the blog post - https://huggingface.co/blog/kalyan-ks/tiny-pii-entity-detection-models
reacted to HannesVonEssen's post with β€οΈ 26 days ago
Post
11631
π£ Hugging Face Visualizer, now as Chrome extension!
https://hfviewer.com
β¨ After installing, Hugging Face model pages will have an architecture visualization on the model page itself!
π Link:
https://chromewebstore.google.com/detail/hugging-face-viewer/mmadlggmpkpiockpjfepaohcllbnakej
Thanks for all the nice feedback so far! β€οΈ
https://hfviewer.com
β¨ After installing, Hugging Face model pages will have an architecture visualization on the model page itself!
π Link:
https://chromewebstore.google.com/detail/hugging-face-viewer/mmadlggmpkpiockpjfepaohcllbnakej
Thanks for all the nice feedback so far! β€οΈ
- 5 replies
reacted to qgallouedec's post with π 26 days ago
Post
10251
Shipped hf-sandbox! π₯‘
π§ͺ Running an eval that executes model-generated C on a few thousand prompts? You probably don't want any of that on your laptop.
Just shipped hf-sandbox, a Modal-style sandbox API on top of Hugging Face Jobs. Spin up an isolated, ephemeral container, run untrusted code, get the result back. No Docker on your laptop, no infra to manage.
Just pip install hf-sandbox.
Early days (v0.1); feedback and issues very welcome:
π https://github.com/huggingface/hf-sandbox
π§ͺ Running an eval that executes model-generated C on a few thousand prompts? You probably don't want any of that on your laptop.
Just shipped hf-sandbox, a Modal-style sandbox API on top of Hugging Face Jobs. Spin up an isolated, ephemeral container, run untrusted code, get the result back. No Docker on your laptop, no infra to manage.
Just pip install hf-sandbox.
Early days (v0.1); feedback and issues very welcome:
π https://github.com/huggingface/hf-sandbox
- 1 reply
reacted to danielhanchen's post with β€οΈ about 1 month ago
Post
8851
We made a guide on how to run open LLMs in Claude Code, Codex and OpenClaw.
Use Gemma 4 and Qwen3.6 GGUFs for local agentic coding on 24GB RAM
Run with self-healing tool calls, code execution, web search via the Unsloth API endpoint and llama.cpp
Guide: https://unsloth.ai/docs/basics/api
Use Gemma 4 and Qwen3.6 GGUFs for local agentic coding on 24GB RAM
Run with self-healing tool calls, code execution, web search via the Unsloth API endpoint and llama.cpp
Guide: https://unsloth.ai/docs/basics/api
reacted to sergiopaniego's post with β€οΈ 2 months ago
Post
2107
TRL is officially an adult π₯³
excited to announce TRL v1.0βοΈ
head to the blog to see how we got here and whatβs next for this post-training library, designed to keep pace with the field
https://huggingface.co/blog/trl-v1
excited to announce TRL v1.0βοΈ
head to the blog to see how we got here and whatβs next for this post-training library, designed to keep pace with the field
https://huggingface.co/blog/trl-v1
- 2 replies
reacted to prithivMLmods's post with π 3 months ago
Post
5067
The Qwen3.5 Multimodal Understanding Demo, powered by Qwen3.5-2B, is now available on HF Spaces! It is a lightweight model designed for fast image and video reasoning. Built with Gradio, the demo showcases Image QA, Video QA, object detection, and 2D point tracking, along with real-time token streaming.
π€ Demo: prithivMLmods/Qwen-3.5-HF-Demo
β Collection: https://huggingface.co/collections/prithivMLmods/multimodal-implementations
π Qwen3.5-2B: Qwen/Qwen3.5-2B
To learn more, visit the app page or the respective model pages.
π€ Demo: prithivMLmods/Qwen-3.5-HF-Demo
β Collection: https://huggingface.co/collections/prithivMLmods/multimodal-implementations
π Qwen3.5-2B: Qwen/Qwen3.5-2B
To learn more, visit the app page or the respective model pages.