Best AI and Machine Learning MCP Servers

Integrates with Penpot's API to enable project browsing, file retrieval, object searching, and visual component export with automatic screenshot generation for converting UI designs into functional code.

Integrates Ollama's local LLM models with MCP-compatible applications, enabling on-premise AI processing and custom model deployment while maintaining data control.

Experience seamless AI-powered 3D modeling by connecting Blender with Claude AI via the Model Context Protocol. BlenderMCP enables two-way communication, allowing you to create, modify, and inspect 3D scenes directly through AI prompts. Control objects, materials, lighting, and execute Python code in Blender effortlessly. Access assets from Poly Haven and generate AI-driven models using Hyper3D Rodin. This integration enhances creative workflows by combining Blender’s robust tools with Claude’s intelligent guidance, making 3D content creation faster, interactive, and more intuitive. Perfect for artists and developers seeking AI-assisted 3D design within Blender’s environment.

Bridges local llama-server instances with MCP clients, providing chat interface, health monitoring, and configurable generation parameters for integrating llama.cpp models with desktop applications

Extracts text from videos and audio files across platforms like YouTube, Bilibili, TikTok, Instagram, Twitter/X, Facebook, and Vimeo using Whisper speech recognition for transcription, content analysis, and accessibility improvements.

MoviePy-based video editing server that provides comprehensive video and audio processing capabilities including trimming, merging, resizing, effects, format conversion, YouTube downloading, and text/image overlays through an in-memory object store for chaining operations efficiently.

Bridges Qwen's code analysis capabilities through CLI integration, providing file-referenced queries with @filename syntax, automatic model fallback, and configurable execution modes for code review, codebase exploration, and automated refactoring workflows.

Integrates with Ollama for local large language model inference, enabling text generation and model management without relying on cloud APIs.

Integrates with n8n workflow automation platform to enable natural language workflow creation, management, and deployment with encrypted credential handling, role-based access control, and automated testing capabilities.

T

Think Tool

Remote

Provides a structured thought process management system for maintaining explicit reasoning steps, policy verification, and tool output analysis through persistent memory storage

Integrates with LlamaIndexTS to provide access to various LLM providers for code generation, documentation writing, and question answering tasks

Integrates with ModelScope's Qwen-Image model to generate images from text prompts, returning base64-encoded results with automatic format detection and error handling for multilingual content creation.

Provides seamless access to OpenRouter's diverse AI models, enabling multimodal interactions across vision and language models with intelligent model selection, caching, and robust error handling.

Integrates with Google's Gemini 2.5 Flash to generate and edit images from text prompts, supporting iterative workflows with reference images and automatic cross-platform file management.

Provides dual-cycle metacognitive reasoning framework that detects when autonomous agents get stuck in repetitive behaviors through statistical anomaly detection and semantic analysis, then automatically diagnoses failure causes and generates recovery strategies using case-based learning.

S

Structured Thinking

Remote

Structures reasoning processes through defined thought stages, managing a history of thoughts with metadata for transparent, step-by-step problem solving and decision making.

Enables semantic search and retrieval of specialized coding skills from GitHub repositories and local directories, using vector similarity to match tasks with relevant bioinformatics, cheminformatics, and data analysis implementations.

Unified server providing access to OpenAI O3, Google Gemini 2.5 Pro, and Azure OpenAI models with automatic usage tracking, cost estimation, and nine specialized development tools for code analysis, debugging, and documentation generation.

Connects AI models to Langfuse analytics workspaces, enabling access to LLM performance metrics by time range for monitoring and analysis.

Tripo 3D

Official

Bridges AI systems with the Tripo 3D generation API, enabling text-to-3D model creation and manipulation through Blender addon integration for seamless 3D content creation workflows.

Integrates with Perplexity API to provide expert programming assistance and persistent chat capabilities for developers and teams.