Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 500+ LLMs (Qwen3, Qwen3-MoE, Llama4, GLM4.5, InternLM3, DeepSeek-R1, ...) and 200+ MLLMs (Qwen2.5-VL, Qwen2.5-Omni, InternVL3.5, Ovis2.5, Llava, GLM4v, Phi4, ...) (AAAI 2025).

Free-Text Promptable Universal 3D Medical Image Segmentation

Open-source temporary elevated access solution for AWS IAM Identity Center.

A framework for few-shot evaluation of language models.

Automatic evals for LLMs

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Synthetic Patient Population Simulator

Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train Qwen3, Llama 4, DeepSeek-R1, Gemma 3, TTS 2x faster with 70% less VRAM.

[DEPRECATED] DICOM WADO Image Loader for the cornerstone library

OHIF zero-footprint DICOM viewer and oncology specific Lesion Tracker, plus shared Meteor packages