Gemini 2.0 Flash was first introduced as an experimental model in December 2024 and became generally available to developers on February 5, 2025. These are not interchangeable dates. The stable `gemini-2.0-flash` endpoint was subsequently shut down on June 1, 2026, according to Google’s official Gemini API deprecation table.
Historical research cutoff: 25 October 2026. Primary-source historical claims are date-scoped; cloud entitlements, region-specific availability, current pricing and rate limits are not implied.
Release identity at a glance
- Provider
- Model family
- Gemini 2.0 Flash
- Public launch
- 2025-02-05
- API ID / checkpoint
- gemini-2.0-flash
- API retirement status
- 2026-06-01
The release and why it mattered
At the December 2024 event Google discussed native tools and potential audio/image output. The developer offering initially gave broad access to multimodal input and text output; some generated media features were restricted to early access. A technically correct release archive must separate research capability, preview availability and stable endpoint support.
Capabilities and genuine use cases
In a production workflow, multimodal inputs allow extraction from mixed documents, images or audiovisual sources, while tool integration can add current data. Neither function excuses an invented citation, and tools must remain subject to application permission checks. Evaluate whether the stable February model actually supports the selected features, rather than reading a demo as a compatibility table.
Integration and version identity
When modernizing a 2.0 Flash app, locate `gemini-2.0-flash` and any `-001` variants, list files and function schemas passed through the API, then retest outputs on a supported model. Logs should distinguish the experimental 2024 ID from the February 2025 stable ID. Google’s shutdown date makes the stable original a historical reference as of cutoff.
Limits and evidence gaps
The Gemini app, AI Studio and Vertex AI have separate availability and control surfaces. Pricing and region claims from a 2025 announcement cannot be carried into a current cost page. The date of shutdown is sourced, but service behavior of every third-party mirror is not asserted.
How to choose or migrate
Gemini 2.0 Flash shows why a model lifecycle needs at least experimental, stable, deprecated and shutdown states. For any new workflow, audit exact endpoint outputs, required modalities and the current API deprecation schedule before selecting a successor.
Reproducible evaluation checklist
For Gemini, explicitly distinguish the Gemini API in AI Studio, the managed Vertex AI surface, and consumer app access. Test preview and stable versions separately with the same multimodal sample set and compare source-grounded extraction rather than marketing accuracy charts.
Before migration, check the official deprecations table for each concrete model ID and note automatic routing aliases. Record inputs, file preprocessing and tool permissions with the request.
Sources & Further Reading
- blog.google — documentation/reference (supplier-provided reference; not independently checked in this pass)
- developers.googleblog.com — documentation/reference (supplier-provided reference; not independently checked in this pass)
- ai.google.dev — documentation/reference (supplier-provided reference; not independently checked in this pass)
Source documentation establishes vendor-published facts, not independently measured model quality, tenant access, contractual rights or current prices. Use a dated pricing ledger and an account-specific entitlement check for deployment decisions.