---
title: "Gemini 2.5 calls re-route on 1 December; Pro users' input price is 20% higher by January"
summary: "Google's Gemini API release notes of 2 November and its price page: where 2.5 Pro and 2.5 Flash calls land, and the per-million-token rates before and after, as of 9 November."
url: https://www.thetechbag.com/blog/gemini-2-5-calls-move-december-price-rise
format: Note
desk: "AI in the enterprise"
region: Global
published: 2026-11-09
facts_checked: 2026-11-09
author: TechBag (The TechBag Journal)
tags: ["Gemini API","Gemini 2.5 Pro","Gemini 2.5 Flash","Gemini 3.8 Flash","Gemini 3.5 Flash-Lite","Gemini 4 Argon","model deprecation","AI pricing","Google AI Studio","Agent Platform"]
cover: https://www.thetechbag.com/blog/covers/gemini-2-5-calls-move-december-price-rise.png
---

# Gemini 2.5 calls re-route on 1 December; Pro users' input price is 20% higher by January

*Google's Gemini API release notes of 2 November and its price page: where 2.5 Pro and 2.5 Flash calls land, and the per-million-token rates before and after, as of 9 November.*

Note · AI in the enterprise · By TechBag · Published 2026-11-09 · Facts checked 2026-11-09 · Canonical: https://www.thetechbag.com/blog/gemini-2-5-calls-move-december-price-rise

From 1 December, every Gemini API request naming `gemini-2.5-pro` will be answered by a different model, Gemini 3.8 Flash, and billed at its rates; requests naming `gemini-2.5-flash` will go to Gemini 3.5 Flash-Lite.[1] For 2.5 Pro users the bill falls first and then partly rises: 3.8 Flash is at an introductory US$0.75 per million input tokens through 31 December and US$1.50 from 1 January, against 2.5 Pro’s US$1.25 for prompts up to 200,000 tokens. Output goes from US$10 to US$3.75, then US$7.50.[2]

## A re-route, not a shutdown

Google’s deprecations table lists 1 December under “Shutdown date”, but the same row says requests “will automatically route” to the replacement.[3] So code that names the old model keeps getting answers, from another model, unless it uses something the new one rejects. Google did the same on 8 October, when calls to 3.7 Flash moved to 3.8 Flash and calls to 3.5 Flash to 3.6 Flash.[1] Access to the 2.5 models has been limited since 18 September to users who had “actively used them in the past”; `gemini-2.5-flash-lite` is not deprecated and stays available “until further notice”.[1][3]

This covers the Gemini Developer API, the one behind Google AI Studio. Google Cloud’s Agent Platform, formerly Vertex AI, runs a separate timetable: its documentation, updated on 7 November, gives 2.5 Pro, 2.5 Flash and 2.5 Flash-Lite a retirement date of 20 October 2026, and the pages we read describe no automatic re-route. It notes that listed dates “refer to retirement of support” on that platform.[4] If your calls go through Google Cloud, check there rather than assume the December date.

## What each call will cost

A separate change arrives the same day. Gemini 4 Argon, released on 13 October, doubles on 1 December from US$2 to US$4 per million input tokens and from US$10 to US$20 for output, the end of a price Google had published with its expiry date. The same rise appears on Agent Platform’s price list.[1][2][5] For 2.5 Flash users, the headline rates do not change: US$0.30 input and US$2.50 output on both models, though audio input falls from US$1.00 to US$0.30.[2]

**Figure: 2.5 Pro users pay less in December; by January input is 20% above today's price, while output stays lower**

*List price, US$ per million tokens, paid tier, standard (not batch): today's rate against the rate after each change*

|  | Today | After the change | Change |
| --- | --- | --- | --- |
| 2.5 Pro input, 1 Dec | 1.25 | 0.75 | -40% |
| 2.5 Pro input, 1 Jan | 1.25 | 1.5 | 20% |
| 2.5 Pro output, 1 Dec | 10 | 3.75 | -62% |
| 2.5 Pro output, 1 Jan | 10 | 7.5 | -25% |
| 2.5 Flash input, 1 Dec | 0.3 | 0.3 | 0% |
| 2.5 Flash output, 1 Dec | 2.5 | 2.5 | 0% |
| 4 Argon input, 1 Dec | 2 | 4 | 100% |
| 4 Argon output, 1 Dec | 10 | 20 | 100% |

Source: Google, Gemini Developer API pricing · as of 9 November 2026 · 2.5 Pro shown for prompts up to 200,000 tokens; above that it costs US$2.50 input and US$15 output

TechBag countWhere the January bill crosses over

On equal token counts, each million input tokens costs US$0.25 more from January and each million output tokens US$2.50 less. The two cancel at ten input tokens per output token. A workload sending 100 million input and 20 million output a month goes from US$325 to US$300; one sending 100 million input and 5 million output goes from US$175 to US$187.50. Prompts above 200,000 tokens, which 2.5 Pro charges at the higher rate, get cheaper at every stage on the published rates. The arithmetic is ours.[2]

## What to test before 1 December

Token counts may not stay equal: thinking tokens are billed as output on both models, and 3.8 Flash’s default thinking level is `medium`.[2][6] Google tells developers to update the model string before 1 December and calls these changes breaking:[1]

- **Thinking settings.** `thinking_level` replaces `thinking_budget`.[6]
- **Sampling parameters.** `temperature`, `top_p` and `top_k` are “deprecated and ignored”, so an application that relies on them loses that control without an error; Google says future generations will return one.[7]
- **Prefilled model turns.** A request ending in a `model` turn returns an HTTP 400 error.[7]
- **Function calling.** Google lists thought signature validation and, on `generateContent`, `call_id` and `name` on every response.[1][6]
- **Grounding allowance.** 2.5 Pro and 2.5 Flash have 1,500 free grounded prompts a day, then US$35 per 1,000. Gemini 3 and later share 5,000 free search requests a month, then US$14 per 1,000, and Google says one request may run more than one search.[2]
- **Output quality.** 2.5 Pro users move from a Pro model to a Flash one, and 2.5 Flash users to a Flash-Lite one. Run your own evaluation set; prices do not measure this.

Google does not say, in what we read, which of these will fail on a re-routed request and which will be tolerated, so test each one against the new model by name.

How we did this

Figures as of 9 November 2026, 7pm IST

Dates and routing are from Google’s release notes and deprecations page for the Gemini API; prices are from its pricing page, standard paid tier, in US dollars, as listed on 9 November. Agent Platform dates are from Google Cloud’s own documentation and differ from the Gemini API’s. Percentages, the break-even and the worked bills are our arithmetic and assume equal token counts, which a change of model may not preserve. We did not call the API, and have not measured how the models compare in output.

Spotted an error? Write to us; corrections are published at the foot of the piece.

Sources

1. [Google: Gemini API release notes](https://aistudio.google.com/docs/changelog) — entries of 18 September, 8 and 13 October and 2 November 2026
2. [Google: Gemini Developer API pricing](https://aistudio.google.com/docs/pricing) — rates dated to 30 November and 31 December 2026, 1 December 2026 and 1 January 2027
3. [Google: Gemini deprecations](https://aistudio.google.com/docs/deprecations) — 2 November 2026 entries
4. [Google Cloud: Gemini Enterprise Agent Platform model versions and Gemini 2.5 Pro](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/model-versions) — last updated 7 November 2026
5. [Google Cloud: Agent Platform pricing](https://cloud.google.com/model-garden/pricing) — rates through 30 November 2026 and from 1 December 2026
6. [Google: What's new in Gemini 3.8 Flash (migration checklist)](https://aistudio.google.com/docs/whats-new-gemini-3.8) — introductory rate through 31 December 2026
7. [Google: What's new in Gemini 3.6 (API changes and parameter updates)](https://aistudio.google.com/docs/whats-new-gemini-3.6) — 2026

**Disclosure.** TechBag is a software reseller: it earns a margin when a reader buys through it, including from vendors named here. The analysis is TechBag’s own; no vendor paid for or saw this piece before publication. Facts are labelled reported, TechBag count or our view.

**Read this next:** [An Agentforce action costs 10 cents, a conversation $2: how AI agents are billed now](https://www.thetechbag.com/blog/how-ai-agents-are-billed-credits-actions) — How five vendors price AI agents by action, conversation, credit, token and seat, from their own price pages, with a month's bill worked through.

Go deeper on TechBag

- [Google Cloud intel page](https://www.thetechbag.com/google/google-cloud) — where Agent Platform, formerly Vertex AI, sits
- [The Google hub](https://www.thetechbag.com/google) — Google Cloud, Workspace and security products
- [Sarvam Prism: input at ₹4 per million tokens](https://www.thetechbag.com/blog/sarvam-prism-decision-model-4-rupees-million-tokens) — per-million-token pricing, in rupees