Skip to content

No UI option to configure agent timeout — local model actions interrupted before completing #3100

@Giri-Aayush

Description

@Giri-Aayush

User GameySins reported on 21/05/2026 that agent actions using a local model are being cut off by a timeout before the model finishes responding. User explicitly wants to extend the timeout — not switch to a smaller/faster model.

Connor's initial response (suggesting the model may be too large for the user's specs) missed the actual request. User clarified they want a configurable timeout setting.

Workaround provided (26/05/2026)
Team advised manually editing the config file (config.yaml, config.json, or .env in the OpenHuman directory) and increasing a timeout value such as action_timeout, execution_timeout, agent_timeout, or request_timeout. Suggested values: 120s (2 min) or 300s (5 min). Docker users should check docker-compose env vars.

This is a FEATURE-REQUEST gap — there is no UI control for timeout settings. Users running large local models on capable hardware have no way to adjust this without editing config files. Resolution unconfirmed — follow-up sent 28/05/2026.


Reported by: @lordam4267
Discord ticket: https://discord.com/channels/1470856511193616498/1507089603356524765
Priority: medium

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions