The official Python SDK for Model Context Protocol servers and clients

A simple screen parsing tool towards pure vision based GUI agent

R1-searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning

Official Repo for Open-Reasoner-Zero

verl: Volcano Engine Reinforcement Learning for LLMs

Super-Efficient RLHF Training of LLMs with Parameter Reallocation

Advancing Language Model Reasoning through Reinforcement Learning and Inference Scaling

Scira (Formerly MiniPerplx) is a minimalistic AI-powered search engine that helps you find information on the internet. Powered by Vercel AI SDK! Search with models like Grok 2.0.

全网乱传的Deepseek从入门到精通的PDF版本，清华大学新闻与传播学院 新媒体研究中心 元宇宙文化实验室

A curated list of reinforcement learning with human feedback resources (continually updated)

Aligning Large Language Models with Human: A Survey

Large-scale, Informative, and Diverse Multi-round Chat Data (and Models)

LLM training in simple, raw C/CUDA

AllenAI's post-training codebase

Keep searching, reading webpages, reasoning until it finds the answer (or exceeding the token budget)

This is a replicate of DeepSeek-R1-Zero and DeepSeek-R1 training on small models with limited data

Fully open reproduction of DeepSeek-R1

Generative Judge for Evaluating Alignment

RewardBench: the first evaluation tool for reward models.

The open-source visual AI programming environment and TypeScript library

Clean, minimal, accessible reproduction of DeepSeek R1-Zero

pingcap/autoflow is a Graph RAG based and conversational knowledge base tool built with TiDB Serverless Vector Storage. Demo: https://tidb.ai

Automatic prompt optimization framework for multi-step agent tasks.

KAG is a logical form-guided reasoning and retrieval framework based on OpenSPG engine and LLMs. It is used to build logical reasoning and factual Q&A solutions for professional domain knowledge ba…

g1: Using Llama-3.1 70b on Groq to create o1-like reasoning chains

Search-o1: Agentic Search-Enhanced Large Reasoning Models