A powerful tool for creating fine-tuning datasets for LLM

tiktoken is a fast BPE tokeniser for use with OpenAI's models.

[Nature Communications] The official code for "Quantifying the Reasoning Abilities of LLMs on Real-world Clinical Cases".

🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!

An interface library for RL post training with environments.

The best ChatGPT that $100 can buy.

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

Open-source implementation of AlphaEvolve

Post-training with Tinker

This is the code repo for the paper AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforced Self-Play (NeurIPS 2025 Spotlight).

Evaluating Multimodal Generative AI with Korean Educational Standards, NAACL 2025.

[EMNLP 2025 Main] CREPE: Rapid Chest X-ray Report Evaluation by Predicting Multi-category Error Counts

Reinforcement Learning with Post-Rollout Edits for Clinically Accurate Chest X-Ray Report Generation

[NeurIPS 2025] This is the official repository for "RAD: Towards Trustworthy Retrieval-Augmented Multi-modal Clinical Diagnosis"

Citrus-V: Advancing Medical Foundation Models with Unified Medical Image Grounding for Clinical Reasoning

slime is an LLM post-training framework for RL Scaling.

Qwen3Guard is a multilingual guardrail model series developed by the Qwen team at Alibaba Cloud.

Project Imaging-X: A Survey of 1000+ Open-Access Medical Imaging Datasets for Foundation Model Development

An Easy-to-use, Scalable and High-performance RLHF Framework based on Ray (PPO & GRPO & REINFORCE++ & vLLM & Ray & Dynamic Sampling & Async Agentic RL)

Training-Ready RL Environments + Evals

Qwen3-omni is a natively end-to-end, omni-modal LLM developed by the Qwen team at Alibaba Cloud, capable of understanding text, audio, images, and video, as well as generating speech in real time.

Think Beyond Images

Understanding Deep Learning - Simon J.D. Prince

🎛 🔊 A Python library for audio.

[NeurIPS 2025] Revisiting End-to-End Learning with Slide-level Supervision in Computational Pathology

CUREBench @ NeurIPS 2025: Benchmarking AI reasoning for therapeutic decision-making at scale

MedAgentBoard: Benchmarking Multi-Agent Collaboration with Conventional Methods for Diverse Medical Tasks

[NIPS 2025] Chiron-o1: Igniting Multimodal Large Language Models towards Generalizable Medical Reasoning via Mentor-Intern Collaborative Search