Gemini 2.5 Pro Drops to Avoid: Google Deprecates 2.5 Pro and 2.5 Flash and Reroutes Their Requests on 1 December
- What happened
- Google deprecated gemini-2.5-pro and gemini-2.5-flash on 2 November. From 1 December 2026 their API requests route to gemini-3.8-flash and gemini-3.5-flash-lite, billed at those models' rates.
- Why it matters
- Any app still calling either id changes model, price and request contract on 1 December, and Google lists breaking changes that can fail requests that work today.
- What to do
- Switch your model string to the replacement and test it before 1 December. We now rate Gemini 2.5 Pro avoid.
Verdict: Gemini 2.5 Pro moves from conditional to avoid. Google deprecated it on 2 November 2026, and from 1 December 2026 a request that names gemini-2.5-pro is answered by Gemini 3.8 Flash and billed at Gemini 3.8 Flash rates. That is 28 days from 3 November. Migrate on your own schedule now, rather than let Google swap the model under a live pipeline.
On 23 October we moved 2.5 Pro to conditional because Google's page said the 2.5 models were "not deprecated" and listed no shutdown date. Both of those facts are now gone, so the verdict moves with them. The same change hits Gemini 2.5 Flash, which is not in our directory.
What happened
Google's Gemini API changelog(opens in new tab), entry dated 2 November 2026:
"
gemini-2.5-prois deprecated and is no longer available to select in Google AI Studio. Effective December 1, 2026, all API requests togemini-2.5-prowill automatically route togemini-3.8-flashand be billed at Gemini 3.8 Flash rates."
The entry for gemini-2.5-flash uses the same wording: requests "will automatically route to gemini-3.5-flash-lite" and are "billed at Gemini 3.5 Flash-Lite rates". Both entries tell developers: "Before December 1, 2026, update your model string" to the new model and review its migration guide.
Google's deprecations page(opens in new tab) now gives both gemini-2.5-pro and gemini-2.5-flash a shutdown date of December 1, 2026, with gemini-3.8-flash and gemini-3.5-flash-lite as the replacements. gemini-2.5-flash-lite still shows "No shutdown date announced".
What this is not: a hard shutdown. Requests to the old ids keep getting answers after 1 December. They just come from a different model, at a different price.
Why it matters
- Your app changes model without a deploy. After 1 December, code that still says
gemini-2.5-progets Gemini 3.8 Flash. Output, latency and token use can all shift, and nothing in your code will show it. - Your bill changes too. Rerouted requests are billed at the new model's rates. We did not read Google's pricing page for this post, so check the current rates for both replacements before you budget.
- Some requests may fail. Google's changelog says: "Breaking API changes include function calling and
thought_signaturevalidation,thinking_levelreplacingthinking_budget, and removed sampling parameters and prefilled model turns." A request that works on 2.5 today can be rejected after the switch. - No more new projects. 2.5 Pro and 2.5 Flash can no longer be selected in Google AI Studio.
What changes for you
| Before (23 October) | Now | |
|---|---|---|
| Rating | Conditional | Avoid |
| Google's status | "Not deprecated", no shutdown date | Deprecated, shutdown date 1 December 2026 |
| Right for | Existing pipelines that already call it | Nobody: migrate before 1 December |
We have not run 2.5 Pro against 3.8 Flash, or 2.5 Flash against 3.5 Flash-Lite, on our own workloads. Test your own prompts on the replacement before you switch, and fix any request that still sets thinking_budget, sampling parameters or a prefilled model turn.
FAQ
Does gemini-2.5-pro stop answering on 1 December? No. Requests keep getting answers, but from Gemini 3.8 Flash, billed at Gemini 3.8 Flash rates.
Is Gemini 2.5 Flash-Lite affected?
Not yet. Google's deprecations page still lists "No shutdown date announced" for gemini-2.5-flash-lite.
Can I still pick 2.5 Pro in Google AI Studio? No. Google says both 2.5 Pro and 2.5 Flash are no longer available to select there.
Gemini 2.5 Pro
Google's Gemini API changelog (2 November 2026) says gemini-2.5-pro is deprecated and no longer selectable in Google AI Studio, and that from 1 December 2026 all API requests to it route to gemini-3.8-flash at Gemini 3.8 Flash rates. The deprecations page lists a 1 December 2026 shutdown date. The 23 October basis (not deprecated, no shutdown date) no longer holds.
What to do
- 1 Search your code and config for gemini-2.5-pro and gemini-2.5-flash, and change them to gemini-3.8-flash and gemini-3.5-flash-lite before 1 December 2026.
- 2 Remove thinking_budget (use thinking_level), sampling parameters and prefilled model turns from requests, and re-test function calling and thought_signature handling, per Google's migration guides.
- 3 Check Google's current rates for Gemini 3.8 Flash and 3.5 Flash-Lite and update your cost budget, since rerouted requests are billed at those rates.
Affected tools & models
Never need to catch up again
The weekly delta: only verdict changes and act-now items. No digest filler.