AI & ML Tools

Open-source embedding database for LLM apps — persistent and in-memory modes.

Official ElevenLabs MCP server for AI-powered text-to-speech, voice cloning, and audio generation.

Unified LLM gateway supporting 100+ providers with a single OpenAI-compatible interface.

A curated MCP toolkit by 21st.dev for rapid product and workflow automation.

Run thousands of open-source AI models (image, video, audio, text) via Replicate's API.

Use Claude models programmatically as tools within your MCP workflow.

Transcribe and analyze audio/video with AssemblyAI's speech intelligence platform.

Query Amazon Bedrock Knowledge Bases using natural language to retrieve relevant information.

Query AWS Bedrock Knowledge Bases for RAG-based document retrieval using natural language from your AI assistant.

Analyze images and videos with AWS Rekognition for object detection, faces, text, and content moderation.

Ultra-fast inference on Cerebras Wafer-Scale Engine chips for LLMs.

Build and deploy visual AI models for image classification, detection, and segmentation via Clarifai.

Generate text embeddings, rerank search results, and run RAG workflows using Cohere's embedding models via MCP.

Access Cohere's language models, embeddings, and rerank APIs from any MCP-compatible client.

Orchestrate Coze's AI bots, workflows, and knowledge bases via the Coze API.

Automate ML model building, deployment, and monitoring with DataRobot's AutoML platform.

Real-time and batch speech recognition powered by Deepgram's Nova-2 models.

Translate text and documents into 30+ languages with DeepL's AI-powered translation engine.

Access DeepSeek reasoning and chat models via MCP for cost-effective AI inference.

Build and invoke LLM apps, chatbots, and agents in Dify's open-source LLM platform.

Generate images via EverArt's AI model fine-tuning platform from your AI assistant with consistent brand style.

Neural search with neural embeddings — find the most relevant web content for any query.

Fastest open-source model inference with function calling on Fireworks AI.

Invoke Flowise AI flows and chatflows with custom inputs via the Prediction API.

Ultra-fast LLM inference via Groq's LPU hardware, exposed as an MCP tool.

Monitor, cache, and rate-limit LLM API calls with Helicone's observability proxy.

Access and run HuggingFace models and Inference API endpoints via MCP.

Manage annotation projects, tasks, and labels in Label Studio for ML training data.

Collaborate on, version, evaluate, and release prompts with Langfuse prompt management.

Trace, debug, and evaluate LLM application runs via LangSmith's observability platform.

Build data-aware AI agents and RAG pipelines using the LlamaIndex framework via MCP.

Run local large language models and expose them via OpenAI-compatible API with LM Studio.

Access geospatial intelligence via Mapbox APIs: geocoding, POI search, directions, and isochrones.

Connect to Mistral AI models (Mistral Large, Mixtral) for chat, completion, and embeddings.

Manage ML lifecycle with MLflow: log runs, register models, and serve predictions.

Deploy and run AI inference functions on Modal's serverless GPU cloud infrastructure.

Run NVIDIA NIM microservices for optimized LLM, vision, and speech inference.

Run local LLMs via Ollama and expose them as MCP tools for offline AI inference.

Use OpenAI Agents SDK tools — web search, file search, computer use, and code interpreter — exposed as MCP endpoints.

Access GPT models, Assistants API, and vector stores via MCP-native interfaces.

Query OpenStreetMap for geocoding, POIs, routes, and geographic data.

Fetch current weather, forecasts, and historical climate data via OpenWeather API.

Prompt optimization server that improves quality and consistency of AI requests.

Train, deploy, and query computer vision models using Roboflow's dataset and inference APIs.

Generate and edit videos with AI using Runway's Gen-3 and other video models.

Submit and manage AI training data tasks through Scale AI's data annotation platform.

Provides structured reasoning helpers for stepwise planning in MCP workflows.

Generate and edit images using Stable Diffusion models via Stability AI's REST API.

Run open-source LLMs (Llama, Mistral, Falcon) at scale via Together AI's cloud API.

Extract and parse text from PDFs, Office documents, HTML, and images for LLM pipelines.

Query Vectara's enterprise RAG platform — hybrid search, grounded generation, and hallucination detection via MCP.

High-throughput open-source LLM serving with PagedAttention, exposed via MCP.

Generate high-quality embeddings optimized for retrieval using Voyage AI's specialized embedding models via MCP.

Track ML experiments, visualize runs, and query model artifacts from W&B.

Transcribe audio files and streams with OpenAI Whisper's state-of-the-art speech recognition.

AI search and coding assistant API with chat, search, and code interfaces from You.com.