SunsetDiff

This model is no longer available to new users

A new Gemini API user can get HTTP 404 on gemini-2.5-flash. Google's deprecations page says access to the 2.5 models is limited to users who have actively used them. gemini-2.5-pro and gemini-2.5-flash are deprecated, with shutdown date December 1, 2026. gemini-2.5-flash-lite is not deprecated.

Error text

This model models/gemini-2.5-flash is no longer available to new users. Please update your code to use models/gemini-3.8-flash for the latest features and improvements. We recommend you to use the Interactions API (https://ai.google.dev/gemini-api/docs/get-started).

Quoted from GitHub issue oyekamal/Design-your-perfect-AI-Persona#1, checked November 3, 2026.

The issue, opened October 31, 2026, prints HTTP 404 with status NOT_FOUND. The message field is the sentence in the Error text block, and it names models/gemini-2.5-flash. The deprecations page does not print that sentence. Checked 2026-11-03, it says access to the 2.5 models is limited to users who have actively used them in the past. gemini-2.5-pro and gemini-2.5-flash are deprecated, with shutdown date December 1, 2026. On that date, requests to gemini-2.5-pro route to gemini-3.8-flash, and requests to gemini-2.5-flash route to gemini-3.5-flash-lite. gemini-2.5-flash-lite is not deprecated and has no shutdown date announced. The error message names models/gemini-3.8-flash. The issue records this sentence for gemini-2.5-flash. It does not record it for gemini-2.5-pro or gemini-2.5-flash-lite. The guide is Gemini 2.5 Pro and Flash auto-route on December 1, 2026. A shorter fragment, from a different issue, is on gemini-2.5-flash is no longer available to new users. A different sentence, recorded for gemini-2.0-flash after that model shut down, is on This model is no longer available. Source: Gemini API deprecations, checked November 3, 2026.

Affected models

The quoted 404 was recorded for gemini-2.5-flash. The other 2.5 ids are the stable models under the same access note. The issue does not record this sentence for them.
Model idWhat the source says
gemini-2.5-flash The GitHub issue records HTTP 404 and this sentence for models/gemini-2.5-flash. The deprecations page lists release date June 17, 2025 and shutdown date December 1, 2026. Requests route to gemini-3.5-flash-lite on that date.
gemini-2.5-pro Release date June 17, 2025. Shutdown date December 1, 2026. Requests route to gemini-3.8-flash on that date. The issue does not record this 404 sentence for this id.
gemini-2.5-flash-lite Release date July 22, 2025. Google says gemini-2.5-flash-lite is not deprecated and remains available until further notice. No shutdown date announced. The issue does not record this 404 sentence for this id.
gemini-3.8-flash The error message says to use models/gemini-3.8-flash. The deprecations note names 3.8 Flash for new projects. The table id on that page is gemini-3.8-flash.
gemini-3.5-flash-lite The deprecations note names 3.5 Flash-Lite for new projects. The table id on that page is gemini-3.5-flash-lite. The error message does not name this id.

Fix

Raw SDK

For a new project, call gemini-3.8-flash or gemini-3.5-flash-lite. The error message names gemini-3.8-flash. On December 1, 2026, a gemini-2.5-flash request routes to gemini-3.5-flash-lite. The guide is the auto-route page. This 404 is the new-user limit.

Before

curl "https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash:generateContent" \
  -H "Content-Type: application/json" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -X POST \
  -d '{"contents":[{"parts":[{"text":"Hello"}]}]}'

After

curl "https://generativelanguage.googleapis.com/v1beta/models/gemini-3.8-flash:generateContent" \
  -H "Content-Type: application/json" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -X POST \
  -d '{"contents":[{"parts":[{"text":"Hello"}]}]}'

Source: Gemini API deprecations, checked November 3, 2026.

LiteLLM

LiteLLM's Gemini API route is the gemini/ prefix plus the model id. For a new project, use gemini-3.8-flash or gemini-3.5-flash-lite.

Python

import litellm
response = litellm.completion(
    model="gemini/gemini-3.8-flash",
    messages=[{"role": "user", "content": "Hello"}],
)

Source: LiteLLM Gemini provider, checked November 3, 2026.

Vercel AI SDK

generateText takes the model id from google(). Use gemini-3.8-flash or gemini-3.5-flash-lite for a new project.

TypeScript

import { google } from "@ai-sdk/google";
import { generateText } from "ai";

const { text } = await generateText({
  model: google("gemini-3.8-flash"),
  prompt: "Hello",
});

Source: AI SDK generateText, checked November 3, 2026.

More request errors

All request errors

Questions

Does Google's deprecations page print this 404 sentence?

No. The page says 2.5 access is limited to users who have actively used those models. gemini-2.5-pro and gemini-2.5-flash are deprecated, with shutdown date December 1, 2026. The sentence in the Error text block is the message field in GitHub issue oyekamal/Design-your-perfect-AI-Persona#1, which records HTTP 404 for gemini-2.5-flash.

Who gets this error?

The sentence says the model is no longer available to new users. The deprecations page says access to the 2.5 models is limited to users who have actively used them in the past, and that new projects should use 3.5 Flash-Lite or 3.8 Flash. The issue records the sentence for gemini-2.5-flash.

Is gemini-2.5-flash shut down?

gemini-2.5-flash and gemini-2.5-pro have shutdown date December 1, 2026. On that date the request routes to another model and the bill changes. gemini-2.5-flash-lite has no shutdown date announced. The guide is Gemini 2.5 Pro and Flash auto-route on December 1, 2026.

Which ids should a new project call?

The error message names models/gemini-3.8-flash. The deprecations note says 3.5 Flash-Lite or 3.8 Flash. The table on that page uses gemini-3.8-flash and gemini-3.5-flash-lite.

Deprecation alerts

We email you before models shut down. Unsubscribe anytime.

Deprecation alert signup
Providers (optional)