Google Will Route Every gemini-omni-1.1-flash Request to Gemini Omni 1.2 Flash on 3 December, Billed at 1.2 Rates
- What happened
- Google made Gemini Omni 1.2 Flash GA on 5 November and will route gemini-omni-1.1-flash requests to it from 3 December, at 1.2 rates.
- Why it matters
- Any app pinned to the 1.1 id changes model and price on 3 December 2026 without a code change.
- What to do
- Switch your model string to gemini-omni-1.2-flash and re-test outputs and costs before 3 December.
Our call: if you pin gemini-omni-1.1-flash, migrate to gemini-omni-1.2-flash yourself before 3 December 2026. Google will move your traffic to Gemini Omni 1.2 Flash on 3 December 2026 whether you have tested it or not, and it will bill you at 1.2 rates from that day.
Affected models: Gemini Omni 1.1 Flash (being retired by routing) and Gemini Omni 1.2 Flash (the new target). Neither is in our directory yet, and we have not tested either.
What happened
The Gemini API changelog entry for 5 November 2026 carries two items that belong together:
- Gemini Omni 1.2 Flash is generally available. Google released
gemini-omni-1.2-flash"featuring up to 15-second video generation and conversational editing", with "30% lower video output pricing ($12.00 per 1M tokens)". - Gemini Omni 1.1 Flash routing notice. "Effective December 3, 2026, API requests to
gemini-omni-1.1-flashwill automatically route togemini-omni-1.2-flashat Gemini Omni 1.2 Flash rates."
The same changelog entry adds two smaller API changes:
- The Gemini API now accepts PPT, ODP and ODS files as input.
- Function responses can now carry audio, video, Office and OpenDocument files.
Why it matters
This is the same pattern Google set for its older text models. From 1 December 2026, Gemini 2.5 Pro requests route to Gemini 3.8 Flash and Gemini 2.5 Flash requests route to Gemini 3.5 Flash-Lite, each billed at the new model's rates.
Silent routing keeps requests working, but they reach a different model at that model's price:
- Price. Google states the 1.2 video output price is 30% lower, at $12.00 per 1M tokens.
- Outputs. A different model can produce different videos, lengths and token counts, so your results and your bill can both move even when the per-token rate drops.
- Timing. If you do nothing, the change lands on Google's date, not after your own regression run.
What changes for you
- Find every call that uses
gemini-omni-1.1-flash, including config files and environment variables. - Switch the model string to
gemini-omni-1.2-flashin a test environment and compare outputs and token usage. - Ship the change before 3 December 2026 so the switch happens on your schedule.
What to do
- 1 Find every call that uses gemini-omni-1.1-flash, including config files and environment variables.
- 2 Switch the model string to gemini-omni-1.2-flash in a test environment and compare outputs and token usage.
- 3 Ship the change before 3 December 2026 so the switch happens on your schedule, not Google's.
Affected tools & models
Never need to catch up again
The weekly delta: only verdict changes and act-now items. No digest filler.