Logo SVG copied to clipboard
Early Access Build agents that never fail Learn more

Run Manifest on your own infrastructure

The full LLM router, self-hosted with Docker. Free and open source, runs offline, and supports local models (Ollama, llama.cpp, LM Studio). One command to get started, no account required.

Quickstart

bash <(curl -sSL https://raw.githubusercontent.com/mnfst/manifest/main/docker/install.sh)

Then open http://localhost:2099 and sign up. The first account becomes the admin. Point your agent at http://localhost:2099/v1 as your OpenAI base URL.

  • Runs on amd64 & arm64
  • Fully offline, no account required
  • OpenAI-compatible API
  • Local models (Ollama, llama.cpp, LM Studio)
    Ollama llama.cpp LM Studio
Docker Hub Image Read the self-hosted docs
ENTERPRISE

Need enterprise features on your infrastructure?

Dedicated managed deployment, SSO, audit logs, uptime SLA, and a support team on your side. Custom pricing for your setup.