The best AI & ML MCP servers
Give your agents the AI and machine-learning building blocks they need — vector databases, embeddings, reasoning tools, and model gateways — as callable MCP tools. Connect once on BusinessMCP and every agent, on any model, works from the same unified endpoint.
Official connectors first, then verified entries, then by how many tools they expose.
Memory
anthropic
Knowledge graph-based persistent memory system for AI. Store and retrieve structured information across conversations.
Chroma
chroma
Open-source vector database for AI applications. Store and retrieve embeddings for RAG and semantic search.
LangChain MCP Adapters
langchain
Bridge between LangChain agents and MCP servers. Use any MCP server as a LangChain tool.
Pinecone
pinecone
Vector database for semantic search and RAG. Store, query, and manage vector embeddings at scale.
Sequential Thinking
anthropic
Dynamic problem-solving through sequential thought processes. Break complex problems into manageable steps.
Math-MCP
EthanHenrickson
Enable LLMs to perform accurate numerical calculations with a simple API. Leverage basic arithmetic and statistical functions.
MindsDB
mindsdb
AI query engine for federated data sources. Run ML predictions through SQL on any database.
DeepSeek
deepseek-ai
Access DeepSeek AI models through MCP. Code generation, reasoning, and general AI assistance.
GitHub Copilot MCP
github
GitHub Copilot integration via MCP. Code completions, explanations, and chat in any MCP client.
Mesh MCP
clay-inc
Access your network seamlessly with a simple and efficient server. Leverage a variety of tools to enhance your applications and workflows.
OpenAI
openai
Access OpenAI APIs through MCP. Use GPT models, DALL-E, Whisper, and embeddings in agent workflows.
Replicate
replicate
Run ML models in the cloud. Access thousands of open-source models for image generation, LLMs, and more.
Weaviate
weaviate
Vector search engine with hybrid search capabilities. Combine vector and keyword search for optimal results.
Hugging Face
Hugging Face
Search the Hugging Face Hub from an agent: find models, datasets and Spaces, read repository details, and browse repository files.
ElevenLabs
elevenlabs
Create, configure and manage the voice and chat agents in an ElevenLabs workspace through the hosted ElevenLabs MCP server.
Bitrix24 MCP-DEV
ExaltedTrou6
MCP server for Bitrix24 provides AI assistants with structured access to the Bitrix24 API for integration development.
Context Awesome
bh-rat
Provide AI agents with instant access to curated resources from over 8,500 awesome lists and more than 1 million items.
Hyperbolic
hyperbolic-labs
Access GPU compute for AI inference. Run open-source models on demand with pay-per-use pricing.
Tailwind Svelte Assistant
CaullenOmdahl
Instant access to SvelteKit and Tailwind CSS documentation along with ready-to-use Svelte component snippets.
Compare all 19 at a glanceShowHide
| Server | Tools | Status |
|---|---|---|
| Memory | 8 | Verified |
| Chroma | 7 | Verified |
| LangChain MCP Adapters | 7 | Archived |
| Pinecone | 7 | Verified |
| Sequential Thinking | 7 | Verified |
| Math-MCP | 8 | — |
| MindsDB | 8 | — |
| DeepSeek | 7 | — |
| GitHub Copilot MCP | 7 | — |
| Mesh MCP | 7 | — |
| OpenAI | 7 | — |
| Replicate | 6 | — |
| Weaviate | 8 | Archived |
| Hugging Face | 4 | Verified |
| ElevenLabs | — | Verified |
| Bitrix24 MCP-DEV | 7 | — |
| Context Awesome | 7 | — |
| Hyperbolic | 6 | — |
| Tailwind Svelte Assistant | 6 | — |
How we rank theseShowHide
Official connectors we host come first, then verifiedentries — ones where we hold the server’s own repository or the vendor’s own docs and have checked the link resolves — then the rest by how many agent-callable tools they expose. A listing on another directory doesn’t count as verification. We don’t publish install counts for third-party servers, because we have no way to measure them, and anything its maintainers have archived says so on the card. A dash in the Tools column means we haven’t listed that server’s tools yet, not that it has none.
Frequently asked questions
What is an AI & ML MCP server?
It exposes AI/ML capabilities — vector search, embeddings, RAG, or model APIs — as agent-callable functions over the Model Context Protocol, so any AI agent can use them without custom integration code.
Can I combine several AI tools in one workflow?
Yes. BusinessMCP unifies every connected server behind one hosted MCP endpoint, so an agent can chain a vector search, an embedding call, and a model query in a single thread.
Do these work across Claude, GPT, and Gemini?
Yes — every server on the registry is model-agnostic. Connect once and the same tools work across all major models and local agents.
Browse other categories
Connect Memory through one endpoint
Reach Memory and the rest of this list from Claude, ChatGPT, Gemini or Cursor without configuring each one. Free plan, no credit card.
Get started freeOr read how managed MCP hosting works first.