Skip to content

feat(llm): add Cheaper Inference configuration template - #110

Merged
cskwork merged 2 commits into
devfrom
feat/cheaper-inference-gateway
Nov 18, 2026
Merged

cskwork merged 2 commits into
devfrom
feat/cheaper-inference-gateway

Conversation

@cskwork

@cskwork cskwork commented Nov 17, 2026

Copy link
Copy Markdown
Owner

Summary

  • Carry the Cheaper Inference template from feat(llm): add Cheaper Inference gateway template #85 onto a branch based on the current dev. The change retains contributor credit and the six-language catalogs and guides.
  • Clear API credentials when a profile crosses the gateway's origin. Prevent stored and active credentials from being reused across that boundary in connection tests and model discovery. Reject an active profile edit that would require a new key.
  • Use the selected Chat Completions output-limit field during the connection test. The connection test keeps its existing diagnostic token budget.

Verification

  • 83 web tests passed and the static build passed.
  • Focused Go tests passed with a disposable PostgreSQL database and loopback HTTP fixtures. No authenticated gateway calls were made.
  • The provider's public model catalog lists gpt-5.4-mini with 400,000 context tokens and 128,000 maximum output tokens.

The shared frontend audit issue in the current base branch is being handled separately.

Keep credentials scoped to their endpoint origin across form edits, connection tests, model discovery, and profile saves. Test the configured Chat Completions token field and preserve the six-language provider guides.

Refs #85

Co-authored-by: aiapienthusiast <327629069+aiapienthusiast@users.noreply.github.com>
@cskwork
cskwork merged commit 04e83e4 into dev Nov 18, 2026
7 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant