Before you read on, three things this is not. It is not an accusation against Google. It does not say anything about every Gemini deployment — only about the route we watched. And it does not replace the incident page itself, which is the record we cannot edit.
Timeline
Timeline and published measurements
- Frozen null floor
- 0.700
- α
- 0.050
- Alarm at
- 20
Incident e3a7e87f · opened 2026-08-05 18:44:17 UTC · published values
| Field | Published value |
|---|---|
| Model / route id | Google Gemini 3.5 Flash Lite google/gemini-3.5-flash-lite |
| Task family | Code reasoning |
| Status at open | Open |
| Opened (UTC) | 2026-08-05 18:44:17 |
| Frozen null floor | 0.700 (α = 0.050) |
| Estimated effect size | 0.633 |
| 95% anytime-valid uncertainty interval | [-0.171, 0.700] |
| E-value at open | 32.365 |
| log10 E at open | 1.510 |
| Onset estimate | 5 (diagnostic only) |
| Serving observation | Via aggregator (not direct adapter) |
| Pool / checker | code_reasoning-v4 / code_reasoning-v2 |
| Incident epoch | 5dad03b2-09b2-49a3-bdf7-ccfaf3543703 |
| Epoch transition | checker_refresh from a4d45cef-6c8c-4239-8d1f-35fc2033c630 |
Validity attaches to the e-process alarm. Default α = 0.05 → threshold 20; opened E = 32.365 crosses that latch.
Route caveat
Aggregator and route caveat
Watchtower records a route: model · upstream · quantisation. Served path via openrouter; upstream/quantisation recorded but not yet pinned. Alarm describes the route as measured.
Observed via aggregator: through an aggregator serving layer, not a direct adapter.
Reference rows never open public incidents by design.
Sibling
Sibling context on the same UTC day
The sibling route gemini-3.5-flash remained quiet on code reasoning. No public alarm appeared on that route. The alarm stayed specific to gemini-3.5-flash-lite.
As of the 2026-08-11 snapshot: quiet on other task families — re-check the live model page before publishing status.
Code reasoning
What "code reasoning" means on this instrument
Deterministic checkers on frozen five-member panels; samples ask for exact stdout of fixed Rust programs. Terminal failure scores 0.
Measured
What was measured; what is not claimed
- Effect size + interval as published
- E-value trajectory
- Quiet elsewhere matters (other families)
- Independence/FAR — methodology; 56.38% block-5 dependence stress
- No customer logos — Actuary’s own public monitor
Why it matters
Why this matters if you do not serve Gemini Flash Lite
CI can stay green while a served aggregator route shifts on one family. Frozen floor + α yields a citable incident. Publishing quiet families avoids selective reporting.
Next step
Next step
Open incident: e3a7e87f-1de1-4212-aab4-8a7342a500c3.