Code-Inc/intelligence-terminal
2
0
Fork 0
Code Issues 11 Pull Requests Actions Packages Projects Releases Wiki Activity

Respect configured LLM timeout and token limits for idea generation #51

New Issue
Closed
opened 2026-07-04 10:09:48 +00:00 by MrSphay · 1 comment
MrSphay commented 2026-07-04 10:09:48 +00:00
Owner
Copy Link

Observed problem

The running LiteLLM integration is reachable and authenticated, but most full sweeps fail with The operation was aborted due to timeout. Small /models and /chat/completions probes return HTTP 200. Some production sweeps have generated 4-5 ideas successfully, confirming intermittent timeout rather than bad credentials.

lib/llm/ideas.mjs hard-codes { maxTokens: 4096, timeout: 90000 }, overriding LLM_MAX_TOKENS and LLM_TIMEOUT_MS configured on OpenAICompatibleProvider/LiteLLMProvider.

Acceptance criteria

  • Idea generation uses the configured provider token and timeout values.
  • Safe fallbacks remain for providers without those properties.
  • Unit tests prove configured values are passed to provider.complete.
  • Gitea runner tests and Docker build pass.
  • A new production image is published and the live Dockge deployment completes a LiteLLM-powered sweep.
## Observed problem The running LiteLLM integration is reachable and authenticated, but most full sweeps fail with `The operation was aborted due to timeout`. Small `/models` and `/chat/completions` probes return HTTP 200. Some production sweeps have generated 4-5 ideas successfully, confirming intermittent timeout rather than bad credentials. `lib/llm/ideas.mjs` hard-codes `{ maxTokens: 4096, timeout: 90000 }`, overriding `LLM_MAX_TOKENS` and `LLM_TIMEOUT_MS` configured on `OpenAICompatibleProvider`/`LiteLLMProvider`. ## Acceptance criteria - Idea generation uses the configured provider token and timeout values. - Safe fallbacks remain for providers without those properties. - Unit tests prove configured values are passed to `provider.complete`. - Gitea runner tests and Docker build pass. - A new production image is published and the live Dockge deployment completes a LiteLLM-powered sweep.
MrSphay commented 2026-07-04 10:12:02 +00:00
Author
Owner
Copy Link

Fix is in PR #52: #52. Live diagnosis confirmed intermittent 90-second timeouts with successful authenticated requests between failures.

Fix is in PR #52: https://git.wilkensxl.de/Code-Inc/intelligence-terminal/pulls/52. Live diagnosis confirmed intermittent 90-second timeouts with successful authenticated requests between failures.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
design-revert
Reverts visual/design changes
 enhancement
New feature or product improvement
 frontend-redesign
Frontend redesign and app-shell/UI work
 marketing-finalization
Product presentation, README, packaging metadata, and customer-facing communication.
 product
User-facing product capability or workflow
 research
Research-originated product idea ready for implementation design
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
MrSphay Toxic
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Code-Inc/intelligence-terminal#51
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?