Skip to content

InferHaven Core v0.1.0

Choose a tag to compare

View all tags
@NimbusSage NimbusSage released this 13 Jun 08:18
· 9 commits to main since this release
v0.1.0
This tag was signed with the committer’s verified signature.
NimbusSage Ethan L.
GPG key ID: C075355AE78EFD6D
Verified
Learn about vigilant mode.
3e2e2b0
This commit was signed with the committer’s verified signature.
NimbusSage Ethan L.
GPG key ID: C075355AE78EFD6D
Verified
Learn about vigilant mode.

InferHaven Core v0.1.0 — first public release

Your own private AI coding server: a self-hostable Docker stack that runs local AI coding assistants on your hardware, with optional GPU acceleration, in a secure terminal-first dev environment. Your models, your code, your box — nothing leaves the machine.

This is the first public release. The project is source-available under FSL-1.1-Apache-2.0 (each version converts to Apache 2.0 two years after release). We're calling it 0.1.0 while the surrounding cloud offering is in beta; 1.0 lands when that stabilizes.

What's in the box

  • Four-service stack on an isolated bridge network: Ollama (OpenAI-compatible inference API), a workspace SSH dev container, code-server (browser VS Code), and Caddy (reverse proxy + automatic HTTPS).
  • The haven CLI — models, background model pulls, per-harness config sync, an always-on auto-restored tmux session, backups (rclone), tmate sharing, GPU readout, and a doctor that checks your whole environment.
  • Seven coding assistants auto-wired to your local models — opencode, aider, qwencode, pi, goose, continue, avante. Pull a model and every harness config re-renders in parallel.
  • GPU or CPU — NVIDIA Container Toolkit for acceleration, or run CPU-only.
  • Devcontainer-ready — VS Code Dev Containers, GitHub Codespaces, DevPod, JetBrains Gateway, headless @devcontainers/cli; two flavors (slim Codespaces + full prod stack), plus nested dev-in-prod.
  • Multi-user, key-only SSH, persistent volumes, BuildKit-cached warm rebuilds < 30s.
  • Optional cloud agent (--profile cloud) — opt-in, zero impact on self-hosted setups.

Quickstart

git clone https://github.com/InferHaven/inferhaven-core.git
cd inferhaven-core
cp .env.example .env        # set CODE_SERVER_PASSWORD + AUTHORIZED_KEYS
make up

Full guide: docs/quickstart.md.

Verify this release

This tag is PGP-signed. Confirm provenance before trusting it:

gpg --import inferhaven_pub.asc        # or from https://inferhaven.com/pgpkey.asc
git verify-tag v0.1.0

Signing key fingerprint: 4992 80D5 D75E 3A4F 837C 6A68 85D8 E097 0D05 CEC0

Security

Found a vulnerability? Please follow SECURITY.md — private disclosure to lookout@inferhaven.com (PGP available), not a public issue.

Contributing

We welcome contributions — start with CONTRIBUTING.md. Non-trivial changes: open a Discussion first.

Assets 2
Loading