llmman: run Claude Code, Codex and other agents on any local model with one command
weskambale · x · 2026-10-30
llmman is a new open-source tool that lets any coding agent run against any model:
- llmman launch claude --model qwen3.8 downloads a llama.cpp build for your GPU, starts a server, loads the model, and points Claude Code at it; Codex, OpenCode and 10+ other agents are supported.
- --provider routes the same agent to OpenRouter, Groq, Anthropic or any models.dev provider, with keys passed per request and never written to agent configs.
- llmman serve speaks Ollama, OpenAI and Anthropic routes on one port; GGUF runs on upstream llama.cpp, safetensors on vllm or mlx-lm.
- Models are plain OCI images pulled from Docker Hub, Hugging Face or your own registry, with optional cosign signing and unsigned-pull rejection. Multiple machines can be aggregated into one endpoint, routing requests to wherever the model is already loaded or has the most headroom.
More from coding & agent
- Pedro Domingos: Claude is 'pretty useless' for code never written before — pmddomingos · 2026-10-30
- AI Code Security Finally Improves: Veracode Benchmark Pass Rate Jumps From 56% to 62% — WeldPond · 2026-10-30
- Scout24's CTO rebuilds org-wide agent workflows on Linear, from idea to merged PR — petergyang · 2026-10-30
- Henge: Open-Source ComfyUI Video Editing Interface Built on Custom MiniMax H3 LoRAs — Round_Awareness5490 · 2026-10-30
- Lean 4 formalization of an asymmetric metric-tensor physics paper hits GitHub — SubtleGradient · 2026-10-30
- YC-backed agent-native office suite launches free in ChatGPT and Codex — ycombinator · 2026-10-30