Manipulating Python Programs

Harbor is a framework for running agent evaluations and creating and using RL environments.

Synthetic data curation for post-training and structured data extraction

Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation

DeepEP: an efficient expert-parallel communication library

FlashMLA: Efficient Multi-head Latent Attention Kernels

json_parser for LLM outputs -> it fixes the malformed json and produces correct json

[NeurIPS 2025 Spotlight] Reasoning Environments for Reinforcement Learning with Verifiable Rewards

Fully open data curation for reasoning models

Scalable RL solution for advanced reasoning of language models

Implementation of the Attention Is All You Need paper

Unsloth Studio is a web UI for training and running open models like Qwen, DeepSeek, gpt-oss and Gemma locally.

A Data Streaming Library for Efficient Neural Network Training

Automatic evals for LLMs

Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.

Clusterfudge examples

Sphynx Hallucination Induction

Enhanced ChatGPT Clone: Features Agents, MCP, DeepSeek, Anthropic, AWS, OpenAI, Responses API, Azure, Groq, o1, GPT-5, Mistral, OpenRouter, Vertex AI, Gemini, Artifacts, AI model switching, message…

An amazing UI for OpenAI's ChatGPT (Website + Windows + MacOS + Linux)

Powerful AI Client

llama3 implementation one matrix multiplication at a time

run paligemma in real time

Tile primitives for speedy kernels

A Continually LoRA PreTrained and FineTuned 7B Llama-2 Indic model for Malayalam Language.

FuseAI Project

OCR, layout analysis, reading order, table recognition in 90+ languages

[ICML'24] Magicoder: Empowering Code Generation with OSS-Instruct