大型語言模型應用

開源大型語言模型應用與框架

這是範圍最廣的一組：聊天介面、SDK 與框架、提示詞與評測工具，以及各種以大型語言模型為基礎的應用。很多專案可以接入多家服務商的模型，有些還能在你自己的硬體上執行模型。

怎麼選

打算用在產品裡的話，先確認三件事：授權——有些熱門專案限制商用；是否支援你正在用的模型服務商；以及它是否還在維護。每個專案頁上的「秒懂」卡片會根據 GitHub 資料回答授權和維護這兩個問題。

星數最多的專案

第 841–864 個，共 866 個

agentlas-ai

Agentlas-OS

★ 1,111

Agent OS: keep specialist agents in a hub, spin up a temporary orchestrator per task. Local-first, works with any model.

Python

nobodywho-ooo

nobodywho

★ 1,111

NobodyWho is an inference engine that lets you run LLMs locally and efficiently on any device.

Rust

berylliumsec

nebula

★ 1,109

AI-powered penetration testing assistant for automating recon, note-taking, and vulnerability analysis.

Python

Onelevenvy

flock

★ 1,109

A desktop multi-agent harness built with Rust, Tauri, and React, powered by langgraph-rust.

Rust

yeahhe365

Gemini-Nexus

★ 1,104

Gemini Nexus 是一款面向浏览器场景的 AI 助手扩展，集成 Gemini Web、Gemini API 与 OpenAI 兼容接口，支持网页上下文、图像处理、工具调用和 MCP 浏览器控制。

JavaScript

EmbeddedLLM

The collaborative spreadsheet for AI. Chain cells into powerful pipelines, experiment with prompts and models, and evaluate LLM responses in real-time. Work together seamlessly to build and iterate on AI applications.

Python

lasgroup

SDPO

★ 1,099

Reinforcement Learning via Self-Distillation (SDPO)

Python

tigicion

dao-code

★ 1,084

Open-source TypeScript terminal coding agent for DeepSeek-V4 — builds on DeepSeek's strong price-performance and ultra-cheap cache pricing, engineering byte-stable prefixes and cache-reusing forks so cross-session memory and a continuous self-correction layer add almost no token cost; 1M context, Skills/MCP/Hooks, Claude Code config compatible.

TypeScript

raindrop-ai

workshop

★ 1,076

Give your coding agent the power to write and run agent evals.

TypeScript

tryAGI

LangChain

★ 1,074

C# implementation of LangChain. We try to be as close to the original as possible in terms of abstractions, but are open to new entities.

OpenCSGs

csghub-server

★ 1,049

此專案圍繞「csghub-server is the backend server for CSGHub which helps user to manage datasets, modes, and also run Model Inference, Finetune and Application Spaces.」建置，聚焦實際場景的開源實作，提供可重用的工具鏈與整合方式。

tomasz-tomczyk

crit

★ 1,034

您與代理商的回饋循環。與您的代理 Claude Code 整合：Crit 還可以與 Cursor、GitHub Copilot、OpenCode、Codex、Gemini、Qwen、Hermes、Windsurf、Cline、Grok、Aider 和 Pi 以及任何可以讀取檔案和運行命令的代理程式配合使用。

Deuz-AI

Deuz-SDK

★ 988

Zero-dependency TypeScript framework for production AI agents: durable execution, long-term memory, hybrid RAG, MCP tool calling, human-in-the-loop approval, planning and CodeAct sandboxes. One streaming API for Claude, GPT, Gemini, Grok, Mistral and DeepSeek — Node, Bun, Deno, serverless and edge.

TypeScript