Official implementation of "Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding"

Implement a reasoning LLM in PyTorch from scratch, step by step

An Open-source RL System from ByteDance Seed and Tsinghua AIR

Minimal yet performant LLM examples in pure JAX

This repo contains the dataset and code for the paper "SWE-Lancer: Can Frontier LLMs Earn $1 Million from Real-World Freelance Software Engineering?"

[NeurIPS 2025 Spotlight] ReasonFlux (long-CoT), ReasonFlux-PRM (process reward model) and ReasonFlux-Coder (code generation)

Reproduce R1 Zero on Logic Puzzle

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

DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models

RAGEN leverages reinforcement learning to train LLM reasoning agents in interactive, stochastic environments.

Janus-Series: Unified Multimodal Understanding and Generation Models

Minimal reproduction of DeepSeek R1-Zero

Simple RL training for reasoning

Fully open reproduction of DeepSeek-R1

DeepSeek-VL: Towards Real-World Vision-Language Understanding

DeepSeek LLM: Let there be answers

The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud.

DeepSeek R1 distilled into smaller OSS models

Friends of OLMo and their links.

A highly capable 2.4B lightweight LLM using only 1T pre-training data with all details.

CleanDiffuser: An Easy-to-use Modularized Library for Diffusion Models in Decision Making

Scalable RL solution for advanced reasoning of language models

An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & TIS & vLLM & Ray & Async RL)

verl: Volcano Engine Reinforcement Learning for LLMs

Let your Claude able to think