• Home
  • Features
  • Pricing
  • Docs
  • Announcements
  • Sign In

lsm / neokai / 25203150941
82%

Build:
DEFAULT BRANCH: dev
Ran 01 May 2026 05:07AM UTC
Jobs 0
Files 0
Run time –
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

Canceled at 05 May 2026 05:05PM UTC via web
25203150941

push

github

web-flow
Fix OpenRouter model cache never refreshing and context window metadata (#1747)

* Fix OpenRouter model cache never refreshing and context window metadata

OpenRouter's modelCache persisted for the process lifetime with no way
to invalidate it. Added clearModelCache(), wired it into the global
clearModelsCache(), and made forceRefresh in models.list actually
synchronous. Also set preferContextWindowMetadata so the API-fetched
context window is used over the SDK default. Added a refresh button
in the Fallback Models settings page.

Co-Authored-By: Claude Opus 4.7 <[EMAIL]>

* Fix forceRefresh to preserve cache on failure; rename to ModelsSettings

refreshModels() only replaces the global cache when the provider fetch
succeeds — a transient API failure no longer nukes the existing model
catalog. Also renamed FallbackModelsSettings → ModelsSettings since it
serves as the primary model catalog page with a refresh button, not
just fallback chain config.

Co-Authored-By: Claude Opus 4.7 <[EMAIL]>

* Remove redundant models.clearCache call from frontend refresh

The daemon-side refreshModels() already handles cache clearing and
safe fallback-on-failure — the frontend calling models.clearCache
first would wipe the global cache and defeat that safeguard.

Co-Authored-By: Claude Opus 4.7 <[EMAIL]>

* Add clearModelCache to Provider interface; extract shared helper

- Added clearModelCache() as optional method on the Provider interface
  so duck-typing checks are unnecessary at call sites
- Extracted clearProviderModelCaches() helper to eliminate the duplicated
  provider-cache loop in clearModelsCache and refreshModels
- Provider-level caches are now only cleared on a full clear (no cacheKey),
  fixing a side effect where per-key clearing wiped all provider caches

Co-Authored-By: Claude Opus 4.7 <[EMAIL]>

* Prevent fallback model data from overwriting richer live cache

When a provider returns static fallback models on API failure (e.g.
OpenRouter returns its... (continued)
Source Files on build 25203150941
Detailed source file information is not available for this build.
  • Back to Repo
  • Github Actions Build #25203150941
  • 306e00e4 on github
  • Prev Build on dev (#25199567865)
  • Next Build on dev (#25218252909)
  • Delete
STATUS · Troubleshooting · Open an Issue · Sales · Support · CAREERS · ENTERPRISE · START FREE · SCHEDULE DEMO
ANNOUNCEMENTS · TWITTER · TOS & SLA · Supported CI Services · What's a CI service? · Automated Testing

© 2026 Coveralls, Inc