AI EngineerGuide

deep-research-pro-preview-12-2025 retirement date: October 23, 2026

Retired on October 23, 2026; requests to deep-research-pro-preview-12-2025 now fail.

Retired on Oct 23, 2026retired
Provider
Google (Gemini API)
Model IDs affected
deep-research-pro-preview-12-2025
Shutdown date
October 23, 2026
Replacement
deep-research-preview-04-2026
Platforms
Dates apply to the Gemini API and are the earliest possible shutdown dates Google publishes. If you use Gemini through another platform, check that platform's own schedule.
Source
Official Google (Gemini API) deprecations page (data last changed 2026-11-02, re-checked daily)

Migrating from deep-research-pro-preview-12-2025

Not yet reviewed. We haven't written model-specific migration notes for deep-research-pro-preview-12-2025 yet. Start with the provider's official guidance and the replacement's model page, and use this general checklist:

  • Search your code, config and environment variables for deep-research-pro-preview-12-2025.
  • Re-run your evaluation set against deep-research-preview-04-2026 before switching traffic.
  • Recount tokens with the new model's tokenizer and re-check max_tokens, context fit and cost.
  • Check tool calling, structured output and stop sequences on your real prompts.
  • Dates apply to the Gemini API and are the earliest possible shutdown dates Google publishes. If you use Gemini through another platform, check that platform's own schedule.

Check your code before Oct 23

Also shutting down on October 23, 2026

All model retirements · All Google (Gemini API) deprecations

FAQ

When was deep-research-pro-preview-12-2025 retired?

deep-research-pro-preview-12-2025 (deep-research-pro-preview-12-2025) was retired on the Gemini API on October 23, 2026, according to the official Gemini API deprecations page. Requests to it now fail.

What should I use instead of deep-research-pro-preview-12-2025?

The recommended replacement is deep-research-preview-04-2026.

What happens to deep-research-pro-preview-12-2025 requests after the shutdown date?

After a model is retired, API requests to it return an error. Move traffic to the replacement and test it on your own prompts before the date.

Does this date apply to Amazon Bedrock, Vertex AI or Azure?

Dates apply to the Gemini API and are the earliest possible shutdown dates Google publishes. If you use Gemini through another platform, check that platform’s own schedule.