Jamba model family — hybrid SSM-Transformer architecture for efficient long-context AI.

High-throughput LLM serving engine — the production standard for GPU inference at scale.

Meta's open-source LLM family — the most popular foundation for self-hosted and fine-tuned AI.

Run 200+ open-source models via fast, affordable API — the one-stop shop for open AI.

The fastest AI inference — custom LPU chips delivering 10x speed for open-source models.

High-performance inference platform — fast, cheap, and optimized for production workloads.

One API, every model — unified access to OpenAI, Anthropic, Google, Meta, and 200+ more.

Run LLMs locally with one command — the easiest way to get AI running on your machine.

Beautiful desktop app for running LLMs locally — discover, download, and chat with AI models.

Open-source ChatGPT alternative that runs 100% offline on your computer.

Free, local, privacy-aware AI — run chatbots on consumer hardware with no GPU required.

Create videos with a text prompt — AI generates script, scenes, voiceover, and music.