gemini-3-flash-preview
gemini-3-flash-preview is a preview model entry of Gemini 3 Flash, suitable for verifying the product experience of low latency, multi-modal interaction and high throughput tasks in advance. It can be used for real-time question and answer, page content understanding, short article generation, material classification and workflow routing in prototypes. The preview phase should pin the model version in the configuration, build a regression test suite, and track response quality, latency, and interface changes. When releasing to official users, it is not appropriate to use the preview capability as the only irreplaceable dependency. A stable version or backup processing link should be prepared.
- Context
- 1048576 tokens
- Max output
- 65536 tokens
- Input
- llm image video text-to-speech
- Output
- llm
Pricing
Input: $0.5 / 1M tokens
Output: $3 / 1M tokens
Cached input: $0.05 / 1M tokens
Audio input: $1 / 1M tokens
Endpoints
- gemini
- openai