Model positioning
Gemini 2.5 Flash is Google's multimodal reasoning model for low-latency, high-throughput tasks. It focuses on balancing performance and cost and is suitable for large-scale processing and agentic applications. (aistudio.google.com)
Capabilities and uses
- Accepts text, image, audio, and video inputs and generates text; supports configurable thinking budgets. (aistudio.google.com)
- Supports function calling, structured output, code execution, context caching, file search, grounding with Google Search and Maps information, and URL context. It can be used for multimodal content understanding, information extraction, batch processing, and agentic workflows. (aistudio.google.com)
- The input limit is 1,048,576 tokens and the output limit is 65,536 tokens; the knowledge cutoff is January 2025. (aistudio.google.com)
Limitations and availability
- This model outputs text and does not directly support image generation, audio generation, or the Live API. Model-generated content may be incorrect, and important results should be independently verified. (aistudio.google.com)
- As of November 18, 2026, Google has deprecated this model, it is no longer available for selection in Google AI Studio, and API access is limited to users who previously actively used the model. Starting December 1, 2026, API requests to
gemini-2.5-flashwill be automatically routed togemini-3.5-flash-liteand billed at the latter's rates; incompatible interface changes must be considered during migration. (aistudio.google.com)
Pricing
The following are the model's prices in US dollars (USD) for the Google Gemini Developer API paid tier, retrieved on November 18, 2026. Except for storage fees, all prices are per 1 million tokens; output prices include thinking tokens. These are not Vertex AI prices. (aistudio.google.com)
| Billing method | Text, image, video input | Audio input | Output |
|---|---|---|---|
| Standard | $0.30 | $1.00 | $2.50 |
| Batch | $0.15 | $0.50 | $1.25 |
| Flex | $0.15 | $0.50 | $1.25 |
| Priority | $0.54 | $1.80 | $4.50 |
For Standard, Batch, and Flex, context cache read prices are $0.03 per 1 million tokens for text, image, and video, and $0.10 per 1 million tokens for audio; cache storage is $1.00 per 1 million tokens per hour. The corresponding Priority read prices are $0.054 and $0.18, respectively, and storage is $1.80 per 1 million tokens per hour. Separate applicable limits and billing rules apply to the free tier and tool calls; see the pricing page for details. (aistudio.google.com)
Release and official sources
- General availability release date: June 17, 2025. Google's announcement states that the generally available version has the same model capabilities as the May 20, 2025 preview. (developers.googleblog.com)
- Official Google release announcement
- Google model documentation
- Google Gemini Developer API pricing
- Google model deprecation notice