run error: LLM request timed out. connected | error agent main | session main (openclaw-tui) | oll

[granite parrot]

i need help

[granite parrot]

help me

[glacial orchid]

Same issue here

[errant valley]

Same here

[fallen moat]

Same first step: what version are you on?

Please paste:

openclaw --version
openclaw status --all

And if the thread title is accurate (LLM request timed out + Ollama), also check the model directly on the gateway machine:

ollama list
ollama ps
openclaw infer model run --model ollama/<your-model> --prompt "reply with ok"

That error usually means the model backend didn’t answer in time — commonly Ollama not running, the selected model being too slow for the machine/context, or a local provider URL mismatch. Docs triage: https://docs.openclaw.ai/help/faq and https://docs.openclaw.ai/help/troubleshooting

[errant valley]

Version 2026.6

How to set the timer longer than the default? For the LLM request

[fallen moat]

For LLM request timed out with Ollama, set the provider request timeout, not the overall agent timeout:

openclaw config set models.providers.ollama.timeoutSeconds 300
openclaw gateway restart

That makes Ollama model HTTP calls wait up to 5 minutes. Docs recommend this first for slow local/remote model servers before raising the whole agent timeout: https://docs.openclaw.ai/providers/ollama / https://docs.openclaw.ai/gateway/local-models

If you’re running the TUI and the whole agent run is timing out instead, that’s separate:

openclaw tui --timeout-ms 300000

But for your specific error text, I’d start with models.providers.ollama.timeoutSeconds. If first-turn model loading is the bottleneck, also consider keeping the model warm with keep_alive in the Ollama model params.

[errant valley]

What if I don't use Ollama but instead an OpenRouter API?

[fallen moat]

Same knob, just under the openrouter provider:

openclaw config set models.providers.openrouter.timeoutSeconds 300
openclaw gateway restart

That extends the OpenRouter model HTTP request timeout to 5 minutes.

Two caveats:

• If you’re using openrouter/auto, the delay may be OpenRouter routing to a slow/downstream model. Picking a concrete model can help:

  openclaw models set openrouter/<provider>/<model>
  

• If the TUI/agent run itself is what times out, also start TUI with a longer outer timeout:

[fallen moat]

  openclaw tui --timeout-ms 300000

Docs: https://docs.openclaw.ai/providers/openrouter and timeout behavior: https://docs.openclaw.ai/concepts/agent-loop

[errant valley]

When I put in - openclaw tui --timeout-ms300000

It gives the error: error: u know option 'timeout-ms300000

[fallen moat]

Tiny syntax thing — the flag needs a space or = before the value:

openclaw tui --timeout-ms 300000

or:

openclaw tui --timeout-ms=300000

--timeout-ms300000 gets parsed as one unknown option name.

But for the LLM request timeout itself, you probably still want the provider timeout too: