An AI-powered research assistant that performs iterative, deep research on any topic by combining search engines, web scraping, and large language models. The goal of this repo is to provide the simplest implementation of a deep research agent - e.g. an agent that can refine its research direction overtime and deep dive into a topic.

Solve Visual Understanding with Reinforced VLMs

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

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

Unleash Next-Level AI! 🚀 💻 Code Generation: DeepSeek r1 + Claude 3.7 Sonnet - Unparalleled Performance! 📝 Content Creation: DeepSeek r1 + Gemini 2.0 - Superior Quality! 🔌 OpenAI-Compatible. 🌊 Streaming & Non-Streaming Support. ✨ Experience the Future of AI – Today! Click to Try Now! ✨

(Supports DeepSeek R1) An AI-powered research assistant that performs iterative, deep research on any topic by combining search engines, web scraping, and large language models.

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。

An open-source alternative to OpenAI and Gemini's deep research.

deepseek java sdk

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

Building DeepSeek R1 from Scratch

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.

Extend OpenRLHF to support LMM RL training for reproduction of DeepSeek-R1 on multimodal tasks.

Explore the Multimodal “Aha Moment” on 2B Model

ReasonFlux: Hierarchical LLM Reasoning via Scaling Thought Templates

Reliable Automation Agents at Scale

推荐使用DeepSeekV3。可以接入微信、QQBot。基于LLM的更逼真的情感陪伴程序。内置了 Atri-My dear moments 的 prompt。Built-in prompt for Atri My dear moments.Support WeChat& QQBot. More realistic emotional companionship program based LLM, meet the characters in your dream.

A one stop shop for data generation, fine tuning and testing LLMs locally using the best tools available. Keeping it simple and versatile!

VimLM is a Vim plugin that provides an LLM-powered assistant for code editing by allowing users to interact with a local LLM model through Vim commands and automatically ingesting code context.

Applying the ideas of Deepseek R1 to computer use

Emacs AI Pair Programming Solution

DeepSeek 逆向 API

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

run DeepSeek-R1 GGUFs on KTransformers

An open-source, high-performance inference accelerator engine supporting various large language models such as DEEPSEEK, GPT, GEMINI, and CLAUDE.

基于 AI 的代码安全审计工具，支持多种编程语言的代码分析，可以帮助开发者快速发现代码中的潜在安全漏洞。支持DeepSeek-R1，ChatGPT-4o等多种大模型。

A CPU Realtime VLM in 500M. Surpassed Moondream2 and SmolVLM. Training from scratch with ease.

基于DeepseekR1/Gemini2/GoogleVison2/GoogleSearch API实现的深度思考联网多模态大模型。

鱼皮的 AI 知识库，汇总热门的 AI 大模型和工具，比如 Deepseek 使用指南、提示词技巧分享、知识干货、应用场景、AI 变现、行业资讯、教程资源等一系列内容，帮助你快速掌握 AI 技术，走在时代前沿。涉及的大模型：chatGPT、Deepseek、Deepseek-r1、QWEN、GROK 等等

DeepSeek Chat UI, Build your own deepseek service through the API of public cloud vendors。支持上下文、历史会话、联网搜索、TXT/DOC/PDF文件上传、用户管理、流式输出。

This repository provides a Python script to fetch and summarize research papers from arXiv using the free Gemini API

Doge Family of Small Language Model

fine-tune deepseek r1

基于200万条医疗数据对DeepSeek-R1-Distill-Qwen-32B进行fine tune且部署

[arXiv] On-device Sora: Enabling Diffusion-Based Text-to-Video Generation for Mobile Devices

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

Local LLM Powered Recursive Search & Smart Knowledge Explorer

TypeScript Agentic AI Library specialized for LLM Function Calling enhanced by compiler skills.

Reflective Planning: Vision-Language Models for Multi-Stage Long-Horizon Robotic Manipulation

探索 AI 的未來！聯發創新基地（MediaTek Research）首次開源的手機應用程式，讓你直接體驗我們最新研發的 AI 模型。透過手機，我們將 AI 科技帶入每個人的生活，完全離線運作，隱私更有保障。這是一個開源專案，我們熱烈歡迎開發者和愛好者一同參與，為 AI 技術發展貢獻力量。立即加入我們，一起打造更優質的 AI 體驗！

使用 Deno 免费代理 Google Gemini，国内直连，不限地区/网络环境，打开即用。

Prompts for deepseek， deepseek R1

DeepSeek 部署教程 / 免费 API 接口 / 官方平替 / 入门指南【2025年3月更新】 提供 DeepSeek 部署教程，包括 PC 电脑端本地部署最新 DeepSeek-R1 模型，手机端通过 Termux 快速部署 DeepSeek-R1 模型。收录各大云平台服务（英伟达、微软、亚马逊AWS、阿里云、腾讯云、硅基流动 x 华为云）提供的（免费） API 接口服务，以及收录支持 DeepSeek API 应用等等。

🔬 Visualize attention layers from Stable Diffusion

Using the moondream VLM with optical flow for promptable object tracking

A step by step guide to fine-tuning the DeepSeek R1 Distilled models on Apple Silicon machines.

Awesome Large Reasoning Model(LRM) Safety.This repository is used to collect security-related research on large reasoning models such as DeepSeek-R1 and OpenAI o1, which are currently very popular.

Local RAG researcher agent built using Langgraph, DeepSeek R1 and Ollama

中文基于满血DeepSeek-R1蒸馏数据集

interact with python in Laravel applications .

SOTA RL fine-tuning solution for advanced math reasoning of LLM

最简易的R1结果在小模型上的复现，阐述类O1与DeepSeek R1最重要的本质。Think is all your need。利用实验佐证，对于强推理能力，think思考过程性内容是AGI/ASI的核心。

[CVPR 2025] Official code repository for "Inst3D-LMM: Instance-Aware 3D Scene Understanding with Multi-modal Instruction Tuning"

VLM^2-Bench: A Closer Look at How Well VLMs Implicitly Link Explicit Matching Visual Cues

An agentic company research tool powered by LangGraph and Tavily that conducts deep diligence on companies using a multi-agent framework. It leverages Google Gemini 2.0 Flash and Chat GPT-4o-mini on the backend for inference.

NativeChat is a powerful application designed to bring seamless, context-aware interactions to your mobile device. Using Gemini under the hood, it offers a wide range of features to assist with tasks like fetching system specs, analyzing call logs, summarizing SMS messages, and providing insights into your device’s status. NativeChat makes your cha

RinAI agentic chat stack. Includes graphRAG memory, parallel tool calling, DeepSeek R1 powered web search, advanced LLM gateway and active context summarization.

A multi-player tournament benchmark that tests LLMs in social reasoning, strategy, and deception. Players engage in public and private conversations, form alliances, and vote to eliminate each other

Notebooks to demo the use of Azure AI Python SDK / LangChain with DeepSeek R1 reasoning model in Azure AI Foundry.

Run DeepSeek R1 Locally with One Command!

An overview of GRPO & DeepSeek-R1 Training with Open Source GRPO Model Fine Tuning

A collection on the recent reproduction papers and projects on DeepSeek-R1

通过Prompt 让deepseek-r1 以及其他模型来支持函数调用

Aims for memory-efficient training (24GB VRAM) on consumer GPUs. Optimizing language models through guidance tokens in reasoning chains, based on DeepSeekRL-Extended.

Enhanced BurpGPT 是一个强大的 Burp Suite 插件。通过分析指定的 HTTP 请求和响应，帮助安全测试人员更快速地发现潜在的安全漏洞。

This repo contains the resources for DeepSeek R1.

Use Stable Diffusion with C#

An AutoHotkey v2 application that utilizes OpenRouter.ai to seamlessly integrate Large Language Models into your daily workflow. Process texts with customizable prompts by pressing a hotkey and interact with multiple AI models simultaneously.

DeepRetrieval - Hacking Search Engines and Retrievers with LLM+RL

DeepSeek Memory is a Python-based interactive terminal application that augments a local Large Language Model (LLM) with persistent memory. It leverages **Ollama** to run a local LLM (e.g., **DeepSeek-R1:1.5b** or any other model you choose) and maintains conversation history across sessions.

🔥🔥🔥Breaking long thought processes of o1-like LLMs, such as DeepSeek-R1, QwQ

🐋 Deepseek Ruby – a simple, easy-to-integrate SDK for accessing the Deepseek API

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

My learning note about DeepSeek-R1 reasoning LLM

本仓库是一个关于 Agent（智能体） 的实验性 DEMO 项目，主要展示如何利用大语言模型（LLM）在计算机上执行“自动化代码编写”、“文件系统探索”等操作。项目基于 DeepSeek R1 系列模型做了一些初步探索，希望借此分享经验与心得，也为更多对 Agent 感兴趣的开发者提供参考。

GRPO for VLMs

A Model Context Protocol (MCP) server implementation connecting Claude Desktop with DeepSeek's language models (R1/V3)

:fire: 安卓手机部署DeepSeek-R1 蒸馏的1.5B模型

🐍 Sublingual helps you log and analyze all of your LLM calls, including the prompt template, call parameters, responses, tool calls, and more. The coolest thing? You don't have to change any of your code!

A repository aimed at pruning DeepSeek V3, R1 and R1-zero to a usable size

Browser based ML Inference | OpenAI compliant | Run models like DeepSeek, Llama 3.2, NomicEmbed, KokoroTTS, and more

仅支持本地ollama deepseek-r1系列模型，同时在POT翻译窗口不显示模型思考过程

A Free AI OpenAI compatible API for Different AI models like GPT-4o, Claude 3.5 Sonnet, Deepseek R1, o3-mini, etc

A curated list of Turkish AI models, datasets, papers

Your AI-powered database assistant. Manage, query, and optimize databases effortlessly with a sleek, Neo Brutalism interface. Chat with your data in real-time, generate reports, and automate complex tasks—all in one place.

Deploying Open-WebUI with Ollama using Docker Compose. Automatically loads deepseek-r1:8b model.

Deepseek-r1复现科普与资源汇总

LLaSE-G1: Incentivizing Generalization Capability for LLaMA-based Speech Enhancement

The objective of this project is to demonstrate how to fine-tune deepseek-r1-distill-llama-8b.

AIDoctor training medical GPT model with ChatGPT training pipeline, implemantation of Pretraining, Supervised Finetuning, RLHF(Reward Modeling and Reinforcement Learning) and DPO(Direct Preferenc…

A benchmarking tool for comparing different LLM API providers' DeepSeek model deployments.