Mastra
81Open-source TypeScript framework for building, running, and observing AI agents in production.
97 verified tools · 14 subcategories
The layer for people building AI: models, inference, vector stores, observability, and GPUs.
Showing 49–72 of 97
Open-source TypeScript framework for building, running, and observing AI agents in production.
Open-source search engine with vector storage and hybrid keyword-plus-semantic retrieval, self-hosted or run as a cloud service
A versioned filesystem built for AI coding agents, adding Git-style branching, checkpoints, and rollback to agent file storage.
On-device AI inference stack for Apple Silicon, aimed at fast, private, local model execution.
French model lab offering open and commercial LLMs, the Vibe assistant, and an EU-hosted API and cloud
A free macOS menu-bar app that reroutes Claude Code and Claude Desktop to run on a different AI model provider.
API of specialized models for coding agents—fast code-apply, code search and context compression—priced per million tokens.
One API over 500+ image, video and audio models, billed per generation with no charge for failed jobs
MIT-licensed JavaScript model that sorts images into porn, hentai, sexy, drawing or neutral in-browser
NVIDIA's open family of AI weather and climate models, covering forecasting, nowcasting, and high-resolution downscaling.
Pay-as-you-go APIs for 200-plus open models, rented GPUs, and a sandbox for running agents.
OpenAI-compatible gateway routing requests to 100+ language, image, and video models by swapping a base URL.
Runs open-weight language models locally from one command, with a hosted tier for larger models
Unified API and agentic router for 400+ LLMs across 100+ providers, billed through one account.
ML platform native to Snowflake that trains, monitors, and retrains models continually instead of from scratch each cycle.
Open-source MCP server that lets AI assistants like Claude Desktop send texts, place calls, and read phone notifications.
Managed vector database for similarity search over billions of embeddings, with serverless and BYOC options
Simulation-driven testing, evaluation, and guardrails platform for AI agents before and after production.
Single async API for image, video, music, chat, and 3D generation models from many providers.
Open-weight coding models, Laguna s and xs 2.1, with an OpenAI-compatible API and a terminal agent
Postgres backend bundled with a RAG pipeline, agent runtime and a visual workflow builder
Runs LLMs and vision models on encrypted GPUs so enterprise data never leaves the customer's environment.
Pay-per-job inference API serving 50+ image and video models with sub-200 ms image latency
Stores and organizes AI prompts across ChatGPT, Claude, Gemini and other models with tags, versioning and sharing.