Gemini 2.0 Flash GA, 2.0 Pro Experimental, 2.0 Flash-Lite
Gemini 2.0 Flash reaches general availability; 2.0 Pro Experimental (2M-token context, strongest coding yet) and cheaper 2.0 Flash-Lite preview ship the same day.
Completes the December 2024 Gemini 2.0 launch. Pro Experimental pairs a 2 million-token window with Search and code-execution tools; Flash-Lite beats 1.5 Flash at the same speed and cost. Text output only at launch; other modalities promised. Flash Thinking Exp-01-21 had shipped 2025-01-21.
- Date
- Wednesday, 5 February 2025
- Lab
- Google DeepMind
- Kind
- model
- Access
- closed API
Figures
| Measure | Value | Measured by |
|---|---|---|
| Context window (2.0 Pro Experimental) | 2M tokens largest Gemini window at the time | company |
Gemini 2.0 models were shut down in the API on 2026-06-01 per the changelog.
Sources
- blog.google/technology/google-deepmind/gemini-model-updates-february-2025/
- ai.google.dev/gemini-api/docs/changelog
This record was checked against its sources on 6 October 2026. How we check