MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
@haisto/vision-mcpgithub.com/haisto/vision-mcpupdated this year
Multi-model vision understanding MCP server. Supports OpenAI-compatible and Anthropic vision models.
npm MIT -
deepseek-vision-mcpgithub.com/kgamv/deepseek-vision-mcpupdated this year
Local-first MCP server for image understanding with Ollama vision models.
MIT ★ 1 -
image-vision-mcpgithub.com/staticdeng/image-vision-mcpupdated this year
Multi-model vision MCP server (GLM-4.6V, Kimi, Qwen-VL, GPT-4o) that lets text-only coding models in Claude Code see and understand images.
npm MIT -
mcp-vision-image-fadhliactive
MCP server untuk analisis gambar lewat 9router, dengan penemuan otomatis model vision yang benar-benar bekerja di mesin Anda. Tanpa quota Google, tanpa daftar model hardcoded.
npm MIT -
vision-mcpgithub.com/forloopcodes/vision-mcpupdated this year
An MCP server that helps non-vision models extract and understand images and screenshots via OpenRouter multimodal models.
MIT ★ 3 -
Computer vision models as MCP servers
MIT ★ 64 -
vision-bridge-mcpgithub.com/mikhailsal/vision-bridge-mcpupdated this year
A lightweight Model Context Protocol (MCP) server for retrieving local and remote images for LLM vision models, featuring metadata extraction and configurable availability retries.
★ 1 -
MCP server and MLOps CLI to export, quantize, audit, benchmark and auto-tune PyTorch and ONNX computer-vision models for edge deployment.
MIT -
percival_vision_mcpgithub.com/bill-kopp-ai-dev/percival_vision_mcpupdated this year
A Model Context Protocol (MCP) server for computer vision.
MIT -
mcp-vision-bridgegithub.com/KuaaMU/mcp-vision-bridgeupdated this year
MCP server that gives text-only LLM coding agents vision — analyze images via any multimodal model (mimo, Claude, Gemini, OpenAI-compatible). Works with Claude Code, Codex, Kimi, opencode, PI.
MIT ★ 13 -
lmstudio-vision-mcpgithub.com/tyfnacici/lmstudio-vision-mcpupdated this year
Use local vision models running in LM Studio from Claude Code through MCP.
MIT -
本项目是一个 **MCP(Model Context Protocol)视觉服务**,核心用途是:**给 Opencode调用的DeepSeek模型配置视觉(Vision)能力**。 由于DeepSeek 模型本身不支持图像输入,因此本项目作为一座"桥":当 DeepSeek(通过 opencode 使用)需要分析图片时,会调用本项目的视觉工具;服务端收到请求后,将图像转发给支持视觉的低成本模型(例如:SiliconFlow 的 Qwen3.5 或Opencode Zen 的 mimo v2.5),再把分析结果返回给 DeepSeek。使得DeepSeek 拥有了"看图"的能力。
MIT -
vision-mcp-servergithub.com/Markusbetter/vision-mcp-serverupdated this year
为你的文字大模型加入一双眼睛。Vision MCP Server adds image understanding to MCP agents by calling a separately configured vision model.
npm MIT ★ 10 -
agent-vision-mcpgithub.com/kitlau86/agent-vision-mcpupdated this year
An MCP server that gives non-vision LLMs the ability to "see" images. Plug in any OpenAI-compatible vision API (Gemini, Qwen-VL, OpenAI, or self-hosted) and your text-only model — like DeepSeek in Claude Code — can analyze screenshots, OCR text, read charts, and more.
MIT ★ 10 -
gimmick-vision-mcpgithub.com/castellotti/gimmick-vision-mcpupdated this year
An MCP server that bridges Claude Code and other agents to any local OpenAI-compatible vision model
MIT ★ 2 -
MCP (Model Context Protocol) server that utilizes the Google Gemini Vision API to interact with YouTube videos.
MIT ★ 6 -
vision-mcp-servergithub.com/gerardodellan/vision-mcp-serverupdated this year
Governed multimodal Vision MCP server — describe/OCR/ask/detect over any OpenRouter vision model; pluggable backend, offline-testable
MIT -
mcp-vision-bridgeupdated this year
Give text-only LLM coding agents (DeepSeek V4 Flash/Pro, Qwen, Kimi) vision via any multimodal model. One analyze_image tool — paste or drag in one or many images, analyzed in a single call, scoped to the current session. Works with Claude Code, opencode,
npm MIT -
nature-vision-mcpgithub.com/fonkychen/nature-vision-mcpupdated this year
An MCP (Model Context Protocol) server that identifies biological species from images using the Nature Vision API, returning Latin names with confidence scores.
npm -
vision-mcpgithub.com/Skirja/vision-mcpupdated this year
MCP server that adds visual understanding capabilities to text-only coding models. Supports OpenAI GPT-4o and compatible vision APIs.
-
vision-mcpgithub.com/weekitmo/vision-mcpupdated this year
MCP server for image understanding through OpenAI-compatible vision APIs. To provide image recognition capabilities for those large models that do not support Multimodal.
MIT -
AI-powered VS Code extension that analyzes Figma designs using vision models (GPT-4o, Ollama) and automatically generates intelligent TypeSpec APIs with dual OpenAPI and MCP specifications.
MIT ★ 1 -
FastMCP server for Moondream vision language model
pypi MIT -
mcp-local-visiongithub.com/jshsakura/mcp-local-visionupdated this year
Give text-only LLMs (GLM, DeepSeek, ...) eyes. MCP server that reads images and PDFs with a vision model on your own GPU — no API keys, nothing leaves the machine.
-
An MCP server for agents that enables models without vision/multimodal support to analyze images using local Ollama vision models.
MIT -
本地视觉理解 MCP 服务,为 AI agent 提供识图能力:通过 vision / ocr / list_models / providers 四个 MCP 工具描述截图、界面、图表与照片。
MIT ★ 5 -
luma-mcpgithub.com/JochenYang/luma-mcpupdated this year
多模型视觉理解 MCP 服务器,为不支持图片理解的 AI 编码模型提供视觉能力:分析截图、报错、UI 与文档,可接入多家主流视觉大模型。Multi-model vision MCP server that adds image understanding to AI coding models without native vision — analyze screenshots, errors, UI and documents via major vision LLM providers.
npm MIT ★ 116 -
her-eyesgithub.com/Penty-d/her-eyesupdated this year
带上她的眼睛 · Give a text-only LLM eyes — a single-binary MCP tool that lets agents like Claude Code / Codex call a vision model to extract structured key information from images.
MIT ★ 2 -
Perceptron MCP server for high-accuracy visual perception powered by fast, efficient vision-language models
npm Apache-2.0 -
@xxtaoaooo-2002/ingredient-analysis-mcpupdated this year
MCP server for analyzing product ingredients using AI vision model
npm MIT -
A token-efficient, provider-agnostic vision layer for reasoning models via MCP.
npm Apache-2.0 ★ 45 -
MCP server that gives text-only coding agents vision — upload images via a binary side channel and analyze them through vision models. Returns structured JSON.
npm MIT -
renderly-mcpgithub.com/xufei547/renderly-mcpupdated this year
MCP server for Renderly — clean web screenshots & content for AI agents (vision models, LLMs, RAG).
MIT -
MCP server for image analysis using OpenRouter's vision models
npm MIT -
visor-mcpgithub.com/ameshkov/visor-mcpupdated this year
MCP server that provides vision tools to text-only models, works with any OpenAI-compatible provider
MIT ★ 9 -
Give text-only LLM agents eyes: an MCP server + CLI that describes images through a vision-language model (VLM) API.
npm MIT -
Official Model Context Protocol (MCP) server for RATA (GHM Automate v2) - Antidetect Browser Automation & AI Vision Engine.
npm MIT -
AI claims triage for car damage: custom-trained vision models as tools for an auditable LLM agent. Models measure, the agent reasons, rules decide.
MIT ★ 3 -
Give text-only LLMs eyes: a proxy swaps images/PDFs for stable text references, and an MCP ask_image tool lets the model question a separate vision model.
MIT -
Desktop + browser GUI agent MCP server. VLM-powered automation: screenshot → vision model → mouse/keyboard/Playwright actions. 9 MCP tools. Fits RTX 4090.
Apache-2.0 ★ 2 -
World perception system combining computer vision models and systems like YOLO, SLAM, Segmentation, VLMs etc into a unified World State Summarizer Interface MCP
MIT ★ 5 -
AdaTile-MCP: high-resolution image adaptive tiling MCP server for DeepSeek vision model (deepseek-v4-flash-vision-exp). L1-L6 pipeline (fastpath, saliency, adaptive tiling, Files API assembly, streaming VLM, rule-based merge) + eval harness. Setup: clone, run setup.bat, add your agent MCP config -> see README.
MIT -
artemotion-mcpupdated this year
Model Context Protocol server for ArtEmotion — full generation catalog (image, video, audio, music, 3d, vision, try-on), AI scaffolder + recommender, server-side pipeline runner, library curation (folders / move / delete), credit + usage tools, job ops (l
npm MIT -
Visual bridge for text-only models and visual enhancement toolkit for multimodal AI agents — OCR, zoom, PDF/Office rendering, web vision and interaction.
★ 20 -
optical-read-mcpgithub.com/hyprcat/optical-read-mcpupdated this year
Read files as densely-packed images instead of text tokens — ~7x fewer tokens for high-res vision models (Claude Fable 5, GPT-5.6 Sol). An MCP server.
MIT -
DeepSeek Harness plugin and OpenCode MCP server: describe an image with any of seven NVIDIA NIM vision models, rotating several API keys. Exposes mcp__opencode_eyes_nvidia__describe_image.
MIT -
Vision models pay for tiles, not megapixels. Per-provider token math, exact + budget resize, zero-dep PNG/JPEG codecs, Claude Code hook, MCP server.
npm MIT -
AI agent that rebuilds 3D CAD solids from 2D technical drawings: a vision model extracts geometric primitives, CadQuery code is generated and run in a sandbox, and the result is checked against bounding box, hole count and volume.
-
adb-mcp-server-progithub.com/tiendung2k03/adb-mcp-server-proupdated this year
A powerful Model Context Protocol (MCP) server for Android control via ADB with self-learning, UI memory, vision-enhanced, and macro support.
MIT ★ 1 -
@iflow-mcp/jochenyang-luma-mcpgithub.com/jochenyang/luma-mcpupdated this year
Multi-model vision understanding MCP server. Supports GLM-4.6V (Zhipu), DeepSeek-OCR (SiliconFlow - Free), Qwen3-VL-Flash (Aliyun), and Doubao-Seed-1.6 (Volcengine)
npm MIT -
Free private photo-to-cartoon and print studio: local computer vision, image analysis, reusable recipes, optional text-model direction and local MCP tools.
MIT ★ 1 -
aitherflow-modelsgithub.com/aitherlab-dev/aitherflow-modelsupdated this year
MCP server for calling external AI models (OpenRouter, Google Gemini, Ollama) with vision support
MIT ★ 2 -
medvision-mcpgithub.com/u9401066/medvision-mcpupdated this year
Medical Vision AI Tools via Model Context Protocol
★ 1 -
MCP server for image analysis using the InternVL3 vision-language model
-
A repository showcasing integration of specialized computer vision tools with language models through MCP.
★ 1 -
omni-mcpgithub.com/Dharit13/omni-mcpupdated this year
Multimodal MCP server — auto-routes to vision, audio, or text models on local Mac M-series hardware
-
Vision Test is an MCP (Model Context Protocol) server that provides a standardized way for AI agents and Large Language Models to interact with mobile devices.
MIT ★ 1 -
gemini-mcp-servergithub.com/deepzhun/gemini-mcp-serverupdated this year
A small, clean MCP server for the Google Gemini API — generate, chat, vision, embed, tokens, models.
MIT ★ 1 -
MCP server for accurate screenshot coordinates with small, fast vision models.
npm MIT -
Official Model Context Protocol (MCP) Server for Tuval Graphic Studio (tuval.site) — Vision-in-the-loop AI Design Automation & Multi-Page PDF Catalogs
npm MIT