A high-throughput and memory-efficient inference and serving engine for LLMs

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

The definitive Web UI for local AI, with powerful features and easy setup.

Topic Modelling for Humans

A next generation HTTP client for Python. 🦋

A hyperparameter optimization framework

min(DALL·E) is a fast, minimal port of DALL·E Mini to PyTorch

View model summaries in PyTorch!

Create Custom LLMs

A neurosymbolic perspective on LLMs

Real programming languages in 50 lines of code

Guaranteed Structured Output from any Language Model via Hierarchical State Machines

TBA