This release includes 1 breaking change for platform teams planning a safe upgrade.
✓ No known CVEs patched in this version
Topics
+9 more
Affected surfaces
ReleasePort's take
Light signalThe release updates the default Gemini model and removes the GEMINI_MODEL environment variable from examples.
Why it matters: If your configuration references GEMINI_MODEL, update it to use the new default model gemini-3.1-flash-lite before deploying version 0.5.2.
Summary
AI summaryRemoved GEMINI_MODEL env var and updated default Gemini model to gemini-3.1-flash-lite.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Feature | Medium |
Updated Gemini default model from `gemini-3.1-flash-lite-preview` to `gemini-3.1-flash-lite` Updated Gemini default model from `gemini-3.1-flash-lite-preview` to `gemini-3.1-flash-lite` Source: llm_adapter@2026-05-21 Confidence: low |
— |
| Deprecation | Medium |
Removed `GEMINI_MODEL` from env examples Removed `GEMINI_MODEL` from env examples Source: llm_adapter@2026-05-21 Confidence: low |
— |
| Refactor | Medium |
Pinned the hosted server to the same Gemini model in code Pinned the hosted server to the same Gemini model in code Source: llm_adapter@2026-05-21 Confidence: low |
— |
Full changelog
Kuku 0.5.2
Small AI model configuration update.
Changes
- Updated Gemini default model from
gemini-3.1-flash-lite-previewtogemini-3.1-flash-lite. - Pinned the hosted server to the same Gemini model in code.
- Removed
GEMINI_MODELfrom env examples.
Self-Hosting Notice
Server deploy is required if you use remote AI. GEMINI_MODEL can be removed from your environment; GEMINI_API_KEY is still required.
Full Changelog: https://github.com/kuku-mom/kuku/compare/0.5.1...0.5.2
Breaking Changes
- Removed `GEMINI_MODEL` environment variable from examples and code.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About kuku-mom/kuku
All releases →Beta — feedback welcome: [email protected]