llm

Dépôts en tendance : LLM

IA & ML

Faire tourner, affiner et servir de grands modèles de langage.

Mis à jour il y a 56 min

Dépôts classés cette semaine
614
Activité par rapport à la semaine dernière
+78 %

Dépôts classés en LLM

100 sur 614 classés · étoiles gagnées cette semaine, rapportées à la taille · min. 20

Comment la santé est calculée
Rang 1

Reverse engineer anything with agents, from app behavior down to native binaries.

TypeScript305 kétoiles au total Santé : En bonne santéTypeScriptAgents IA
+230 k étoiles cette semaine
Rang 2

Let an AI model use a real iPhone with a $7 ESP32: AirPlay eyes, USB HID hands, any OpenAI-compatible vision model. No Mac, no developer account, no jailbreak.

C3 kétoiles au total EmbarquéC & C++
+3 k étoiles cette semaine
Rang 4

基于 REA 深度拆解并高保真还原 Claude 官方桌面端 UI。还原 30 个独立 React 组件与 8 项硬核交互物理算法,附带 3000+ 行逆向分析手册。

TypeScript1,1 kétoiles au total Rétro-ingénierieUI & design systems
+1,1 k étoiles cette semaine
Rang 5

Autonomous AI sales agent powered by Claude Code or Codex. Finds prospects, writes cold emails, sends campaigns, handles replies, books meetings. No API keys needed — runs on your Claude or Codex subscriptions.

Python1,2 kétoiles au total Santé : En bonne santéClaude CodePython
+1,1 k étoiles cette semaine
Rang 7

Scout: the open-source browser agent, by Outcrawl. Mirrored automatically from Outcrawl's own repository.

TypeScript933étoiles au total Tests & QATypeScript
+933 étoiles cette semaine
Rang 8

The operational superset of Pi Coding Agent — everything Pi, plus observability, governance, recovery, evaluation and multi-agent orchestration. Pi Coding Agent 的运维超集——完整继承 Pi 的核心能力,并进一步扩展可观测性、治理、故障恢复、评估和多 Agent 编排。

TypeScript1 kétoiles au total ObservabilitéApplis de bureau
+927 étoiles cette semaine
Rang 9

AI API 中转站评测排行与切换工具:实测可用率、延迟与价格 · Open-source AI API relay benchmark, rankings & switcher for Claude Code / Codex

Rust1,4 kétoiles au total Claude CodeApplis de bureau
+1,2 k étoiles cette semaine
Rang 10

Unlock Google 11's Gemini "Rambler" Natural Voice Dictation & AI Writing Tools on ANY Android phone. No root required.

Java948étoiles au total JavaAndroid
+822 étoiles cette semaine
Rang 11

Every agent's model. One place. Codex on DeepSeek, Claude Code on Kimi, from the menu bar.

Go15 kétoiles au total GoClaude Code
+6,4 k étoiles cette semaine
Rang 12

Intent compiler for AI agents — converges vague requests into typed IntentSpec contracts (probe, ask, or halt before routing), the input layer for routers and typed-decision models like Jev & Laya

Python763étoiles au total Skills d’agentsClaude Code
+692 étoiles cette semaine
Rang 13

Local-first AI teammate workspace. Bring your own models and keep project conversations, tool approvals, and results together.

TypeScript620étoiles au total Santé : En bonne santéLocal-firstMCP
+618 étoiles cette semaine
Rang 14

Local gateway for Claude Code, Codex and other AI clients on macOS, Windows and Linux: switch upstreams without touching clients, keep API keys from relays, cut off malicious tool calls, trace every request.

TypeScript837étoiles au total WindowsClaude Code
+702 étoiles cette semaine
Rang 15

AI-assisted RAW photo developer: GPT vision advisor proposes an EditRecipe, a deterministic Rust engine renders it — Lightroom-compatible XMP sidecars, 24 RAW formats + baked images, measured (not guessed) Lightroom mask geometry, local GUI/web UI, AI denoise & segmentation sidecars

Rust1,2 kétoiles au total Santé : En bonne santéVision par ordinateurRust
+830 étoiles cette semaine
Rang 16

Open-source personal finance app with an AI financial advisor: track your net worth, investments, ETFs, cash and debts on your own computer.

Python4 kétoiles au total Auto-hébergéPython
+2 k étoiles cette semaine
Rang 17

CodeAgent is an open-source coding agent that lives in your terminal. It understands your repository, plans multi-step changes, writes and edits files, runs commands and tests, and iterates until the task is done — all with a transparent, extensible tool loop.

Python865étoiles au total TerminalOutils CLI
+647 étoiles cette semaine
Rang 18

For You Agent——AI 时代的个人随身数字人格。把你的模型、AI 账号、技能、提示词和工作方式,带到每一个 AI 工具里。

Rust933étoiles au total Santé : En bonne santéSkills d’agentsApplis de bureau
+684 étoiles cette semaine
Rang 19

Official repository for MOSS-Image, a 5B-parameter base DiT model with high-quality text rendering, layout generation, and efficient fine-tuning.

Python422étoiles au total Vision par ordinateurPython
+422 étoiles cette semaine
Rang 20

llama.cpp on a jailbroken PS5: run any GGUF model natively on the console, starting with Qwen 3.8

C++487étoiles au total LLM en localmacOS
+414 étoiles cette semaine
Rang 21

Project Maya - run GLM-5.3-Flash (321B MoE) privately on your own GPU(s), from one card to 16, NVIDIA or AMD, Linux or Windows. Engine, server, dashboard, OpenAI- and Anthropic-compatible API, quant tools.

C++1 kétoiles au total LLM en localAuto-hébergé
+699 étoiles cette semaine
Rang 22

37 tok/s from Qwen3-Next-80B (48 GB) on a 32 GB laptop, 24 without drafts vs 2.8 for llama.cpp; reading 62 vs 16 tok/s. Faster than llama.cpp on all 13 models compared in Q4_K_M. LLMs bigger than your RAM, pure C, CPU or GPU.

C431étoiles au total LLM en localC & C++
+399 étoiles cette semaine
+1 autres de Jakeschincariol
Rang 82

Eleven free Claude skills that run a LinkedIn account: posts off 21 hook formulas, comments, replies, profile score, weekly plan, and a humanizer that strips the AI fingerprint and scores the draft before it goes out.

Python2,3 kétoiles au total Skills d’agentsClaude Code
+520 étoiles cette semaine
Rang 24

DeepSeek Harness 破甲:让所有模型都能破甲,不同模型可换不同提示词;默认提示词面向国模「小码酱」。Jailbreak for every model — swap prompts per model. 求 Star 收藏 ⭐ 如要二开,请查看署名要求部分

JavaScript6,7 kétoiles au total JavaScriptAgents de code
+2,4 k étoiles cette semaine
Rang 26

Better judgment. Better taste. Less ceremony. System instructions and source-aware steering for GPT in Codex.

JavaScript351étoiles au total JavaScriptAgents de code
+351 étoiles cette semaine
Rang 27

Fully autonomous AI Agents system capable of performing complex penetration testing tasks

Go30,9 kétoiles au total Santé : En bonne santéSécuritéGo
+5,5 k étoiles cette semaine
Rang 28

☤ The official plugins for Hermes Agent, built and maintained by Nous Research: Home Assistant, Spotify, Claude Subscription DirectSDK and more, one install away from the Hermes plugin catalog.

Python314étoiles au total DomotiquePython
+310 étoiles cette semaine
Rang 29

Evolving Multimodal Inference Programs: LLM-guided evolutionary search for accurate and efficient vision-language inference.

Python332étoiles au total Python
+296 étoiles cette semaine
Rang 30

Claudemy: an open-source alternative to Udemy and Coursera. Give Claude a topic, it researches the web and builds a course of 2-minute audio lessons with cited sources. A free Claude Code skill.

Python295étoiles au total Claude CodePython
+295 étoiles cette semaine
Rang 31

Describe any machine in one line. Claude builds a photoreal, physics based, interactive simulation of it in one HTML file, plus the film to post it. An open source Claude Code skill.

JavaScript290étoiles au total Skills d’agentsJavaScript
+290 étoiles cette semaine
Rang 32

A growing collection of viral videos made with Claude Opus 5.5 and the prompts behind them. Watch each original next to a live remake on Skillry. Updated regularly.

5,7 kétoiles au total
+1,8 k étoiles cette semaine
Rang 34

Find the cheapest API for any model and connect it from your AI agent: Grandet Skill and an anonymous MCP server, backed by listed prices from 1,800+ official, inference and relay providers on grandet.ai.

JavaScript276étoiles au total MCPSkills d’agents
+270 étoiles cette semaine
Rang 35

Context window optimization for AI coding agents. Sandboxes tool output (98% reduction), persists session memory, and enforces routing across 17 platforms via MCP + hooks.

TypeScript30,5 kétoiles au total Santé : En bonne santéMCPAgents de code
+4,1 k étoiles cette semaine
Rang 36

Claude Code skill for short films with no video model: 51 film styles, each a style prompt plus a demo film made entirely in code by Claude Opus 5.5 (incl. OPUSCAR 98 and the new Evidence Films). By Lemomo. · Opus5.5 x 51 种影片风格:风格提示词 + 纯代码样片 + 导演与技术指南(新增「实证影像」)

JavaScript2,9 kétoiles au total Skills d’agentsJavaScript
+1,1 k étoiles cette semaine
Rang 38

Build AI agents that run 100% on-device. Sub-100ms latency on Qualcomm NPU. Zero cloud dependency.

C++1,1 kétoiles au total Santé : Signaux mitigésEmbarquéC & C++
+549 étoiles cette semaine
Rang 39

Claude Code, Codex, Copilot, Pi, OpenCode, Gemini, and Prime Agent versions of Poteto's pstack. Rigorous agent workflows with Cursor primitives translated for other harnesses.

JavaScript2,7 kétoiles au total Santé : En bonne santéSkills d’agentsJavaScript
+971 étoiles cette semaine
Rang 40

Make Gemini code like Claude Models. Forces it to inspect files before editing, plan changes, test its work, and back up every “done” with actual evidence.

Python239étoiles au total Skills d’agentsPython
+230 étoiles cette semaine
Rang 41

Claude Code plugin to see, review and iterate on your agent's work without leaving the terminal: show images, play videos, and run visual QA with Playwright recordings and before/after comparisons inside the Claude Code terminal. Works in any terminal that supports the kitty graphics protocol, fully cross-platform on macOS, Linux and Windows.

TypeScript229étoiles au total TerminalTests & QA
+228 étoiles cette semaine
Rang 42

nanoMuse: an open-source personal agent for every device you own. One agent with a name and a look of its own that does things, keeps working while the app is closed, remembers you, and asks before anything you cannot undo. On Android, iPhone and iPad, Windows / macOS / Linux and in the browser; your phone's screen and your computer as its hands.

Swift1,4 kétoiles au total SwiftiOS
+613 étoiles cette semaine
Rang 43

🛡️ Free open-source AI-powered security terminal & vulnerability scanner (CVE, SBOM). Supports SSH, SFTP, RDP, VNC, Serial, and 12+ autonomous AI agents (DeepSeek, OpenAI) for automated security workflows, CTF & DevSecOps. Cross-platform & Web UI.

HTML830étoiles au total TerminalSécurité
+424 étoiles cette semaine
Rang 44

Frametell: turn your ideas and material into vivid explainer videos with an AI director agent. Local-first, any model.

TypeScript208étoiles au total Vision par ordinateurLocal-first
+208 étoiles cette semaine
Rang 45

Can AI engines cite your site, and do they? Free 0–100 readiness score on an open GEO rubric, plus ask and weekly citation tracking via OpenAI, Perplexity, Gemini, Anthropic and OpenRouter with your own keys. Zero dependencies; CLI, GitHub Action and MCP server.

Python324étoiles au total Outils CLIMCP
+213 étoiles cette semaine
Rang 46

7.4 billion tokens per month. 34 free LLM providers. 635 free model endpoints. All behind one /v1 endpoint, plus any custom OpenAI-compatible endpoint. Smart routing, automatic failover, encrypted keys. Personal experimentation only.

TypeScript36,5 kétoiles au total Santé : En bonne santéTypeScript
+3,7 k étoiles cette semaine
Rang 48

Fold papers, web pages, files, and AI replies into five levels of detail. Zoom from an overview to the original text.

Python336étoiles au total Local-firstPython
+215 étoiles cette semaine
Rang 50

Interactive desktop app with 51 labs covering ML fundamentals, transformers, RAG, agents, fine-tuning and quantization and more.

TypeScript196étoiles au total Machine learningLocal-first
+193 étoiles cette semaine
Rang 51

ARTEX 한국어판 · AI 자율 침투 테스트 프레임워크(upstream: Autumn-27/ARTEX, AGPL-3.0)

Go1,6 kétoiles au total SécuritéGo
+608 étoiles cette semaine
Rang 52

A comprehensive logo-design skill for Claude, Gemini CLI, Codex and other AI agents: principles, process, SVG craft, testing tools and a 1,400+ logo reference library.

HTML3,6 kétoiles au total Skills d’agentsAgents de code
+946 étoiles cette semaine
Rang 53

Free SEO MCP server + open-source SEO and GEO skills for Claude: keyword research, rank tracking, audits, backlinks, AI visibility on your real GSC/GA4/ads data. claude mcp add ryze --transport http https://connector.get-ryze.ai/mcp

Shell5,9 kétoiles au total MCPClaude Code
+1,3 k étoiles cette semaine
Rang 54

A super fast Graph Database uses GraphBLAS under the hood for its sparse adjacency matrix graph representation. Our goal is to provide the best Knowledge Graph for LLM (GraphRAG).

Rust10,4 kétoiles au total Santé : En bonne santéBases de donnéesRust
+1,7 k étoiles cette semaine
Rang 55

Walk through a diff the way a senior engineer reads it. A read-only TUI that turns a git change into a guided tour.

Go176étoiles au total Outils CLITerminal
+176 étoiles cette semaine
Rang 56

Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 250,000+ scientists worldwide. 177 ready-to-use validated skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigravity, and the open Agent Skills standard.

Python52,1 kétoiles au total Santé : En bonne santéBases de donnéesSkills d’agents
+3,9 k étoiles cette semaine
Rang 58

AirLLM 2.8T LLM inference with single 4GB GPU

Jupyter Notebook39,1 kétoiles au total Santé : Signaux mitigés
+3,3 k étoiles cette semaine
Rang 59

Harness self-evolution: update targets, theory, evaluation, and reported evidence. Bilingual reading list and project website.

TeX168étoiles au total Skills d’agentsAgents de code
+166 étoiles cette semaine
Rang 60

Autonomous GitHub Project build → QA → review loop for Claude Code. One-line install.

HTML357étoiles au total Santé : Signaux mitigésTests & QAAgents de code
+204 étoiles cette semaine
Rang 62

Your Claude Code, Codex and Mistral Vibe sessions as a live, animated team call: who is working, who waits on whom, who needs you. VS Code extension, terminal command, desktop app. Runs locally, zero tokens.

HTML235étoiles au total Éditeurs & IDEObservabilité
+163 étoiles cette semaine
Rang 65

腾讯 CodeBuddy 账号池管理控制台,内置 OpenAI 兼容反代网关:扫码批量纳管账号、定时签到与保活、密钥分组分发、IP 与模型白名单、调用日志与用量统计;支持带签名校验的一键更新。

Python1,4 kétoiles au total RéseauUI & design systems
+491 étoiles cette semaine
Rang 66

Answer me with HTML — an agent skill that answers hard questions with a one-page HTML you can actually read. 让 AI Agent 用一页 HTML 回答复杂问题。

JavaScript3,6 kétoiles au total Outils CLISkills d’agents
+832 étoiles cette semaine
Rang 68

Clone any viral video with AI agents. Not just a script, the whole workflow: swap the face, the words, the B-roll, ship 100 variants in one command, and get your 100M views.

TypeScript23 kétoiles au total Santé : En bonne santéVision par ordinateurTypeScript
+2,2 k étoiles cette semaine
Rang 69

JiuwenSwarm is an intelligent AI Agent built on openJiuwen. It extends the powerful capabilities of large language models directly to your fingertips through various communication apps you use daily.

Python7,1 kétoiles au total Santé : En bonne santéAgents IAPython
+1,2 k étoiles cette semaine
Rang 71

Free open-source extractor for AI coding assistant chat histories. Supports Claude Code, Cursor, Windsurf, Aider, Cline/Roo Code and more

Python151étoiles au total Agents de codeClaude Code
+150 étoiles cette semaine
Rang 72

免 ROOT 免 Termux,在手机上跑 DeepSeek Harness。完整 Ubuntu 环境 + proroot 零 ptrace 开销 · AI 输出实时上屏 · ADB 直连 · 数据不丢

Java1,5 kétoiles au total Santé : En bonne santéJavaAndroid
+469 étoiles cette semaine
Rang 73

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

JavaScript166 kétoiles au total Santé : En bonne santéSkills d’agentsJavaScript
+5,8 k étoiles cette semaine
Rang 74

Connect your users to their apps

TypeScript15,2 kétoiles au total Santé : En bonne santéMCPJavaScript
+1,6 k étoiles cette semaine
Rang 75

Run large AI models on your own PC: Qwen3.8-Flash-Next on Strata's engine, and GLM 5.3 Flash, DeepSeek V4 Flash, LFM2.5, Qwen3.6, Ornith 1.5, Gemma 4, Tiel Coder and MiniCPM-V on a second engine written for PolyStrata. One-click install for Windows and Linux, OpenAI/Anthropic API on localhost.

C++188étoiles au total LLM en localAuto-hébergé
+139 étoiles cette semaine
Rang 76

Claude Code skill for Dribbble-level product launch videos from code. One shape that never cuts, springs, a cursor that drives every change, beat-synced SFX, 60 fps motion blur. v2: use with Claude Opus 5.5.

JavaScript301étoiles au total Santé : Signaux mitigésTests & QAVision par ordinateur
+163 étoiles cette semaine
Rang 77

AI stickman animation style library & video prompts.|火柴人动画画风大全:输入话题,生成脚本、动态分镜及配音音效提示词。

Python2,2 kétoiles au total Santé : Peu d’activitéSkills d’agentsPython
+557 étoiles cette semaine
Rang 78

Open decision models from 0.8B to 27B: a calibrated probability for every option in one pass, and a short thought only when unsure. Engine for vLLM, llama.cpp and MLX.

Python619étoiles au total LLM en localAgents IA
+256 étoiles cette semaine
Rang 79

Tamper-evident tracing, policy gates and cross-checks for AI agents: see what the agent did, and why.

Python134étoiles au total ObservabilitéMCP
+131 étoiles cette semaine
Rang 80

An open-source take on ChatGPT's Intelligent UI: answers stream as interactive UI (maps, photos, charts, drawings and working tools). Built with OpenUI.

TypeScript521étoiles au total Frameworks webTypeScript
+225 étoiles cette semaine
Rang 83

Mods for Claude Code. effortless picks the reasoning effort for every prompt and shows the cache countdown. textless (desktop app) shows art, quotes and chess instead of the wall of text while Claude works.

HTML230étoiles au total Applis de bureauClaude Code
+126 étoiles cette semaine
Rang 84

19 open-source MCP servers (308 tools) that give AI agents the Iranian web: Digikala, Snappfood, Jabama, MrBilit, Paziresh24, Taaghche and more. Prices in Toman.

Python122étoiles au total Skills d’agentsMCP
+122 étoiles cette semaine
Rang 85

The open-source app everyone uses to manage agents at work

TypeScript104 kétoiles au total Santé : En bonne santéSkills d’agentsMCP
+3,8 k étoiles cette semaine
Rang 86

A command-line tool for Z-Library. Search, download, send to Kindle, and integrate with AI agents.

Go989étoiles au total Santé : Signaux mitigésTerminalOutils CLI
+313 étoiles cette semaine
Rang 87

Official codebase for Spotter: Let the Embodied Model Lead, and the VLM Reflect for It

Python202étoiles au total RobotiqueAgents IA
+119 étoiles cette semaine
Rang 88

Turn Claude into a senior design architect: DTCG tokens, 52 components, WCAG 2.2 AA-AAA, 138 design systems, any-framework code, and 50 objective gates that fail the build instead of faking success.

JavaScript2 kétoiles au total Santé : Signaux mitigésUI & design systemsFrameworks web
+465 étoiles cette semaine
Rang 89

Long-term memory runtime for AI agents — plain Markdown as the source of truth, local ranked retrieval, and an independent sleep-time Manage layer. Claude Code and Codex share one store. No API key.

Python4,1 kétoiles au total Bases de donnéesLocal-first
+678 étoiles cette semaine
Rang 90

Google Ads MCP server + Meta Ads MCP (Facebook Ads MCP) + Google Analytics MCP (GA4) + Search Console in one hosted remote MCP for Claude, ChatGPT, Cursor & n8n: OAuth login, no API keys, approval-gated writes, 7-day free trial. By Ryze AI.

5,1 kétoiles au total Santé : Peu d’activitéMCPAgents de code
+770 étoiles cette semaine
Rang 91

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

JavaScript282 kétoiles au total Santé : En bonne santéMCPJavaScript
+6 k étoiles cette semaine
Rang 92

Turn any idea, plan, or codebase into a beautiful interactive diagram. An agent skill for Claude Code, Codex, and more.

JavaScript84,8 kétoiles au total Santé : En bonne santéDatavizSkills d’agents
+3,2 k étoiles cette semaine
Rang 93

Agent + reusable skill library for LIBERO manipulation tasks (no RL / VLA / world model; no LLM calls at runtime)

Python143étoiles au total RobotiquePython
+112 étoiles cette semaine
Rang 94

World's first open-source, agentic video production system. 12 production pipelines, 100+ tools, 700+ agent skill and production-knowledge files. Turn your AI coding assistant into a full video production studio.

Python69,1 kétoiles au total Santé : Signaux mitigésVision par ordinateurSkills d’agents
+2,9 k étoiles cette semaine
Rang 95

A tiny friend in your Mac's notch and on your iPhone that keeps an eye on your AI coding agents: Claude Code, Codex, Cursor, Gemini CLI, Antigravity and more. Approve from the notch or your Lock Screen.

Swift5,4 kétoiles au total SwiftiOS
+749 étoiles cette semaine
Rang 96

📡 Your own AI-powered news radar. Generates daily briefings in English & Chinese. | 用 AI 构建你专属的新闻雷达

Python10,7 kétoiles au total Santé : En bonne santéMCPPython
+1,1 k étoiles cette semaine
Rang 98

一个自己找热点、自己写日报的网站框架。把信源和精选标准换成你的,它就是你的行业热点站。

TypeScript7,9 kétoiles au total Bases de donnéesAuto-hébergé
+864 étoiles cette semaine
Rang 99

DeepSeek Harness(DSH)一只住在 DSH 界面右下角的小鲸鱼娘,帮你盯着DeepSeek账户余额。QQ弹弹,支持拖拽吸附、左吸附翻转、数字滚动动画,随界面自动启用,建议直接喊来你的dsh安装

JavaScript5,1 kétoiles au total Santé : En bonne santéJavaScriptAgents de code
+683 étoiles cette semaine
Rang 100

TokenSpeed is a speed-of-light LLM inference engine.

Python2,7 kétoiles au total Santé : En bonne santéPython
+480 étoiles cette semaine