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

Scalable toolkit for efficient model reinforcement

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

The lightweight framework for building agents

A clean, modular SDK for building AI agents with OpenHands V1.

Deep Agents is an agent harness built on langchain and langgraph. Deep Agents are equipped with a planning tool, a filesystem backend, and the ability to spawn subagents - making them well-equipped…

Use Claude Code as the foundation for coding infrastructure, allowing you to decide how to interact with the model while enjoying updates from Anthropic.

Nano vLLM

A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.

Ongoing research training transformer models at scale

τ²-Bench: Evaluating Conversational Agents in a Dual-Control Environment

Renderer for the harmony response format to be used with gpt-oss

Kernel sources for https://huggingface.co/kernels-community

Train transformer language models with reinforcement learning.

Load compute kernels from the Hub