Free AI tools for AI Infrastructure & Dev Platforms

Meta's family of open-weight LLMs, downloadable under a community licence with a 700M-MAU carve-out.

German non-profit publishing open image-text datasets like LAION-5B, plus CLIP models, for ML research

Open-source visual builder for LLM agents and RAG flows, deployable as an HTTP API or an MCP server.

Builds an open-source, Postgres-based memory engine that gives AI agents relational, vector, and graph retrieval.

A free macOS menu-bar app that reroutes Claude Code and Claude Desktop to run on a different AI model provider.

NVIDIA's open family of AI weather and climate models, covering forecasting, nowcasting, and high-resolution downscaling.

Find where to submit your startup: MCP server + planner

Open-source app and library for running, fine-tuning and reinforcement-learning open LLMs locally with far less GPU memory.

Free WordPress plugin that runs an MCP server inside the site, so Claude, ChatGPT or Cursor can research, write and publish posts.

Cloud data lake plus no-code ML workspace, with time-series forecasting and deployable REST endpoints.

Open-source desktop chat app that runs open models offline or connects to cloud providers, plus a terminal coding agent.

Open-source MCP server that lets AI assistants like Claude Desktop send texts, place calls, and read phone notifications.

Free platform for optimizing, quantizing and deploying PyTorch or ONNX models to run on Qualcomm phones, PCs, cars and IoT chips.

Query language and runtime for constrained LLM prompting, built at ETH Zurich's SRI Lab.

MIT-licensed Python/JavaScript framework for building LLM agents, with 1,000+ model and tool integrations

Self-hosted open-source runtime that serves text, speech, vision, image and agent workloads behind OpenAI-compatible APIs.

Open-source platform for tracking ML experiments, registering models and tracing, evaluating and deploying LLM agents.

MIT-licensed JavaScript model that sorts images into porn, hentai, sexy, drawing or neutral in-browser

Open-source deep learning framework for building and training neural networks in Python, with GPU acceleration and a large ecosystem.

Open-source machine learning framework for building, training and deploying models on servers, browsers, phones and edge devices.

Open-source Python toolkit that scores RAG and agent apps with LLM-based metrics and generates synthetic test sets.

Google DeepMind's family of open-weight language and multimodal models, from phone-sized to workstation-sized, free to download and run.