Gemini 2.5 Pro and Flash deprecated, rerouted from December 1
Requests switch to newer models and their prices on December 1. Breaking API changes come with the move.
Google deprecated gemini-2.5-pro and gemini-2.5-flash on the Gemini API on November 2, 2026. From December 1, 2026, requests to either model ID will be routed to a newer model and billed at that model's rates.
What changed
- From December 1, requests to
gemini-2.5-proroute togemini-3.8-flashand are billed at Gemini 3.8 Flash rates. - From December 1, requests to
gemini-2.5-flashroute togemini-3.5-flash-liteand are billed at Gemini 3.5 Flash-Lite rates. - Both models can no longer be selected in Google AI Studio. Google says API access to the 2.5 models is limited to users who have actively used them in the past.
- Google lists breaking API changes for both moves: function calling and
thought_signaturevalidation,thinking_levelreplacingthinking_budget, and removed sampling parameters and prefilled model turns. gemini-2.5-flash-liteis not deprecated and has no shutdown date.
Google prices
Gemini API pricing page, paid tier, standard, November 3, 2026, per million tokens. Gemini 2.5 Pro prices are for prompts up to 200K tokens. Gemini 2.5 Flash prices are for text, image and video input.
| Model | Input | Output |
|---|---|---|
| gemini-2.5-pro | $1.25 | $10.00 |
| gemini-3.8-flash, through Dec 31, 2026 | $0.75 | $3.75 |
| gemini-3.8-flash, from Jan 1, 2027 | $1.50 | $7.50 |
| gemini-2.5-flash | $0.30 | $2.50 |
| gemini-3.5-flash-lite | $0.30 | $2.50 |
What it means
Gemini 2.5 Pro traffic gets cheaper in December at Gemini 3.8 Flash's current rates. From January 1, 2027, 3.8 Flash input costs $1.50, more than 2.5 Pro's $1.25, while output stays below 2.5 Pro's $10. Gemini 2.5 Flash traffic moves to a model with the same list price. In both cases a different model answers under the old ID, and requests that use thinking_budget, sampling parameters or prefilled model turns need code changes.
Gemini on RouterPlus
Gemini 2.5 rows and the Flash-Lite replacement in the RouterPlus catalog, November 3, 2026, per million tokens:
| Model | Provider | Input | Output |
|---|---|---|---|
| Gemini 2.5 Pro | DeepInfra | $1.25 | $10.00 |
| Gemini 2.5 Pro | Inference.net | $1.25 | $10.00 |
| Gemini 2.5 Flash | DeepInfra | $0.30 | $2.50 |
| Gemini 2.5 Flash | Inference.net | $0.30 | $2.50 |
| Gemini 3.5 Flash Lite | Inference.net | $0.30 | $2.50 |
| Gemini 2.5 Flash Lite | Inference.net | $0.10 | $0.40 |
Gemini 3.8 Flash is not in the RouterPlus catalog. Google's notice covers its own Gemini API and does not say what other hosts will do. Compare every model on the RouterPlus models page.
What to do
Update the model string on the Gemini API before December 1:
gemini-2.5-pro -> gemini-3.8-flash
gemini-2.5-flash -> gemini-3.5-flash-lite- Run your own test prompts on the replacement first, using Google's migration guides.
- Replace
thinking_budgetwiththinking_level, and remove sampling parameters and prefilled model turns. - If you move 2.5 Pro traffic to 3.8 Flash, budget for its January 1 price change.