OpenHands

All Hands. Docs checked 2026-10-12. Docs currently name Qwen3.6-35B-A3B as a local starting point.

Software-agent UI/CLI that talks to any LiteLLM-supported model, including local OpenAI-compatible servers.

What it’s for

Multi-step coding agent in a browser or CLI. Needs a capable instruction-tuned model.

Local endpoint

llama-server base: http://127.0.0.1:8080/v1
Strata base: http://127.0.0.1:8083/v1
Model name: openai/<id from GET /v1/models> (the openai/ prefix is required)
API key: LLM_API_KEY=local-llm (any placeholder if the server has no auth)

Tool calling

Agent tools via the OpenHands stack. Local models must be strong at tool-following — not documented as guaranteed on every GGUF.

Long context

Not documented as a first-class slider; size -c on the server.

ROCm / Radeon gotchas

OpenHands does not ship ROCm. You own llama-server / vLLM. Docker setups must use host.docker.internal, not 127.0.0.1.

Hybrid local + cloud

Yes — cloud LiteLLM providers plus local LLM_BASE_URL.

This is the same question as the calculator’sfully localversushybridtoggle. This page does not quote cloud prices.

Official links

Sources

Copy-paste

More on the connect page. Prove http://127.0.0.1:8080/v1/models first.

OpenHands → llama-server :8080environment

# Docs: LLM_BASE_URL + openai/<id> prefix. Docker must use host.docker.internal.
export LLM_BASE_URL=http://127.0.0.1:8080/v1
export LLM_API_KEY=local-llm
export LLM_MODEL=openai/qwen

Exact llama-server flags for our daily driver stay TODO in stacks.yaml. This is the port only.

OpenHands → Strata :8083environment

# Docs: LLM_BASE_URL + openai/<id> prefix. Docker must use host.docker.internal.
export LLM_BASE_URL=http://127.0.0.1:8083/v1
export LLM_API_KEY=local-llm
export LLM_MODEL=openai/qwen

Exact Strata flags for our rig stay TODO in stacks.yaml. This is the port only.