AI agents running research on single-GPU nanochat training automatically

Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

A character-level language diffusion model trained on Tiny Shakespeare

The best ChatGPT that $100 can buy.

TPU inference for vLLM, with unified JAX and PyTorch support.

EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL

PyTorch-native post-training at scale

Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthr…

Our library for RL environments + evals

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

LLM KV cache compression made easy

Rust-tokenizer offers high-performance tokenizers for modern language models, including WordPiece, Byte-Pair Encoding (BPE) and Unigram (SentencePiece) models

A framework for few-shot evaluation of language models.

A Lightweight LLM Post-Training Library

A library to train, evaluate, interpret, and productionize decision forest models such as Random Forest and Gradient Boosted Decision Trees.

Library for reading and processing ML training data.

A JAX-native High Performance Eval Metrics Library

A simple, performant and scalable Jax LLM!

📋 A list of open LLMs available for commercial use.

Multi-backend recommender systems with Keras 3

You like pytorch? You like micrograd? You love tinygrad! ❤️

Rax is a Learning-to-Rank library written in JAX.

Bringing BERT into modernity via both architecture changes and scaling

A collection of LLM papers, blogs, and projects, with a focus on OpenAI o1 🍓 and reasoning techniques.

Build resilient language agents as graphs.

High-performance, asynchronous Python HTTP client library designed for faster file transfers using concurrency, semaphores, and fault-tolerant features.