#663 · Primary category: AI Agents & Automation

helix

agents api genai glm golang helm k8s kimi llm llm-agent llm-serving openai openapi qwen rag self-hosted swagger swarm

♾️ Private Agent Fleet with Spec Coding. Each agent gets their own GPU-accelerated desktop. Run Claude, Codex, Gemini and open models on a full private AI Stack ♾️

Project last updated:08/29/26

GitHub Stars

803

Forks

83

Contributors

34

License

Other

Why we included this project

Helix is for teams that are tired of running one coding agent per terminal on a developer's laptop and want the whole thing on shared infrastructure instead. It runs many agents in parallel, each in its own isolated GPU-accelerated desktop with a browser, terminal, and filesystem, and you steer the work from a spec-driven Kanban board: you approve a plan, the agent implements it, and you review the pull request it opens. The harnesses are swappable per task, so you can use Claude Code, Codex, Gemini CLI, Qwen Code, or any ACP-compatible agent, and point them at hosted models or self-hosted ones on your own GPUs, even air-gapped. RAG, tracing, multi-tenancy, and billing are built in, which makes it closer to a complete private GenAI stack than a thin wrapper. If your bottleneck is babysitting one agent in one terminal, this gives you a team-level workflow with review gates instead.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category