An open source deep research clone. AI Agent that reasons large amounts of web data extracted with Firecrawl

Open Source Deep Research Alternative to Reason and Search on Private Data. Written in Python.

A fast inference library for running LLMs locally on modern consumer-class GPUs

Agent Laboratory is an end-to-end autonomous research workflow meant to assist you as the human researcher toward implementing your research ideas

LLaSA: Scaling Train-time and Inference-time Compute for LLaMA-based Speech Synthesis

HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal

Universal and Transferable Attacks on Aligned Language Models

Next-Token Prediction is All You Need

Janus-Series: Unified Multimodal Understanding and Generation Models

Scalable toolkit for efficient model alignment

Structured Text Generation

Distribute and run LLMs with a single file.

Implementation for MatMul-free LM.

A large-scale information-rich web dataset, featuring millions of real clicked query-document labels

Stanford NLP Python library for Representation Finetuning (ReFT)

LDB: A Large Language Model Debugger via Verifying Runtime Execution Step by Step

OpenChat: Advancing Open-source Language Models with Imperfect Data

Tools for merging pretrained large language models.

Implementation of MusicLM, Google's new SOTA model for music generation using attention networks, in Pytorch

Python bindings for llama.cpp

MLX: An array framework for Apple silicon

Fixes mojibake and other glitches in Unicode text, after the fact.

Generative Representational Instruction Tuning

Implementation of DoRA

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

Finetune Llama 3.3, DeepSeek-R1, Gemma 3 & Reasoning LLMs 2x faster with 70% less memory! 🦥

DSPy: The framework for programming—not prompting—language models

SOTA Google's Perceiver-AR Music Transformer Implementation and Model