Top Open Source AI Organization Influencers in 2025Q1

Fully open reproduction of DeepSeek-R1

The Open All-in-One Multimodal AI Agent Stack connecting Cutting-edge AI Models and Agent Infra.

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

A community-driven AI automation framework that builds upon the incredible work of the open source community. Our goal is to combine language models with specialized tools for tasks like web search, crawling, and Python code execution, while giving back to the community that made this possible.

Solve Visual Understanding with Reinforced VLMs

A high-performance LLM inference API and Chat UI that integrates DeepSeek R1's CoT reasoning traces with Anthropic Claude models.

YuE: Open Full-song Music Generation Foundation Model, something similar to Suno.ai but open

Sim Studio is an open-source AI agent workflow builder. Sim Studio's interface is a lightweight, intuitive way to quickly build and deploy LLMs that connect with your favorite tools.

Witness the aha moment of VLM with less than $3.

🐬DeepChat - A smart assistant that connects powerful AI to your personal world

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

SpatialLM: Training Large Language Models for Structured Indoor Modeling

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

The official repo of MiniMax-Text-01 and MiniMax-VL-01, large-language-model & vision-language-model based on Linear Attention

Skywork-R1V2:Multimodal Hybrid Reinforcement Learning for Reasoning

A collection of production-ready Generative AI Agent templates built for Google Cloud. It accelerates development by providing a holistic, production-ready solution, addressing common challenges (Deployment & Operations, Evaluation, Customization, Observability) in building and deploying GenAI agents.

A fast Rust based tool to serialize text-based files in a repository or directory for LLM consumption

Open Source AI coding assistant for planning, building, and fixing code. We're a superset of Roo, Cline, and our own features. Follow us: kilocode.ai/social

Run LLMs with MLX

An AI chat bot for small and medium-sized teams, supporting models such as Deepseek, Open AI, Claude, and Gemini. 专为中小团队设计的 AI 聊天应用，支持 Deepseek、Open AI、Claude、Gemini 等模型。

RikkaHub is a Android APP that supports for multiple LLM providers.

VS Code extension for LLM-assisted code/text completion

Awesome RL Reasoning Recipes ("Triple R")

A simple open-source chat app that uses Exa's API for web search and Deepseek R1 for reasoning

LLM/VLM gaming agents and model evaluation through games.

deepseek java sdk

Bytebot is the container for desktop agents.

TypeScript AI Framework specialized AI Function Calling enhanced by compiler skills.

🌐 WebThinker: Empowering Large Reasoning Models with Deep Research Capability

Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models

Explore the Multimodal “Aha Moment” on 2B Model

An open-source tool for seamless migration from other LLMs to Llama, and for general prompt optimization.

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

A-MEM: Agentic Memory for LLM Agents

LM Studio Python SDK

ReasonFlux Series - Open-source innovative LLM post-training algorithms focusing on data selection, reinforcement learning, and inference scaling

⚡️ supercharged community-maintained PHP API client that allows you to interact with deepseek API

Build your own visual reasoning model

Flame is an open-source multimodal AI system designed to translate UI design mockups into high-quality React code. It leverages vision-language modeling, automated data synthesis, and structured training workflows to bridge the gap between design and front-end development.

Universal AI agent building apps using Amazon Bedrock, capable of customize to create/edit files, execute commands, search the web, use knowledge base, use multi-agents, generative images and more.

Laravel wrapper for Deepseek PHP client, to seamless deepseek API integration with laravel applications.

Reliable Automation Agents at Scale

GPT-4o-level, real-time spoken dialogue system.

Notate is a desktop chat application that takes AI conversations to the next level. It combines the simplicity of chat with advanced features like document analysis, vector search, and multi-model AI support - all while keeping your data private.

A lightweight reproduction of DeepSeek-R1-Zero with indepth analysis of self-reflection behavior.

Applying the ideas of Deepseek R1 to computer use

Simple extension on vLLM to help you speed up reasoning model without training.

Doge Family of Small Language Model

🌊 Simple, event-driven and stream oriented workflow for TypeScript

🧠🔗 Graph-Based Programmable Neuro-Symbolic LM Framework - a production-first LM framework built with decade old Deep Learning best practices

BreezeAPP 是一款為 Android 和 iOS 平台開發的純手機 AI 應用程式。從 App Store下載，即可在不連網的狀態下享受多項 AI 功能。源碼由聯發創新基地(MediaTek Research)提供。我們旨在推廣兩個概念: 人人都可以在自己的手機上自由選擇並運行不同的LLM - one is free to choose one's own LLM to run on a phone，以及任何app開發者都可以輕鬆寫作創意的純手機AI應用 - any dev can create purely phone-based AI apps easily。

LLMVoX: Autoregressive Streaming Text-to-Speech Model for Any LLM

Go (Golang) client for Deepseek API. Deepseek Go supports DeepSeek-V3, DeepSeek-R1 and more

Decentralized LLMs fine-tuning and inference with offloading

Dive is an open-source AI Agent desktop application that seamlessly integrates any Tools Call-supported LLM with frontend MCP Server—part of the Open Agent Platform initiative.

A token management platform that reverse-engineers the conversation interfaces of ChatGPT, Cursor, Grok, Claude, Windsurf, Gemini, and Sora, converting them into the OpenAI format./Token管理平台，逆向ChatGPT、Cursor、Grok、Claude、Windsurf、Gemini、Sora平台的对话接口转OpenAI格式

Community maintained hardware plugin for vLLM on Ascend

Code for the paper "Vamba: Understanding Hour-Long Videos with Hybrid Mamba-Transformers" [ICCV 2025]

[SIGGRAPH 2025] ReStyle3D: Scene-level Appearance Transfer with Semantic Correspondences

Prompts for deepseek， deepseek R1

Demonstrates how to build a streaming enterprise agent using Azure AI Agent Service. This sample integrates local HR and company policy documents, Bing for external context, and GPT-4o (though the service is model-agnostic) to deliver real-time Q&A and automated workflows.

Make your LLM agent and chat with it simple and fast!

This is the repo of developing reasoning models in the specific domain of financial, aim to enhance models capabilities in handling financial reasoning tasks.

A user-friendly Command-line/SDK tool that makes it quickly and easier to deploy open-source LLMs on AWS

FakeVLM: Advancing Synthetic Image Detection through Explainable Multimodal Models and Fine-Grained Artifact Analysis

Collection of demos for building Llama Stack based apps on OpenShift

MoD Control Tile Upscaler for SDXL Pipeline

AI-powered computer control for automated testing. Factifai uses vision models (Claude, GPT-4o, Gemini) to interact with applications naturally - clicking, typing, and verifying results just like a human would.

Demonstrate how to use DeepSeek-R1 models GitHub Models, Azure or Ollama

ScalarLM - a unified training and inference stack

Delightful AI coding mentor | Open-source, local, hackable | 🌟 Star if it sparks delight!

AIN - The First Arabic Inclusive Large Multimodal Model. It is a versatile bilingual LMM excelling in visual and contextual understanding across diverse domains.

Official implementation of LangCoop: Collaborative Driving with Natural Language

Dremio MCP server

Official implementation for "Diffusion Instruction Tuning"

Linq's Function Calling Agent utilizing DeepSeek-R1

ChinaTravel: A Real-World Benchmark for Language Agents in Chinese Travel Planning

A framework making it effortless to convert any llm model into a reasoning agent like o1 or DeepSeek's r1

A novel alignment framework that leverages image retrieval to mitigate hallucinations in Vision Language Models.

智谱 glm realtime api python/golang/ts sdk, 包括 low level 的 websocket client 封装以及各个场景的调用样例

Multi-modal agentic framework for surgical procedures

Convert LLaMA3.1-8B to DeepSeek R1 MLA & MoE (raw)

FastLLM - Rust based LLM Inference API

DeepSeek-R1 JavaScript starter

用于astrbot，使用Stable Diffusion WebUI的API进行图片生成的插件。

SpringAI，LLM，MCP，Embedding

SQL GPT agent that understands schema. Runs queries and displays results from natural language.

Reproduction of the complete process of DeepSeek-R1 on small-scale models, including Pre-training, SFT, and RL.

minRAG is a RAG system that starts from scratch, pursuing the ultimate simplicity and power. It consists of no more than 10,000 lines of code, requires no installation, and can be launched with a double-click

LM Studio App and Developer Docs

Ops files for https//github.com/meta-llama/llama-stack

强大的浏览器插件，支持deepseek r1直接以侧边栏的形式和页面进行交互。

[ICLR 2025] - Cross the Gap: Exposing the Intra-modal Misalignment in CLIP via Modality Inversion

Safety Mirage: How Spurious Correlations Undermine VLM Safety Fine-tuning

Hana is an AI agent built with the Zen package.

An LLM-powered chatbot for fediverse. A tech demo for BotKit.