gemini-3.1-pro-preview
gemini-3.1-pro-preview is a preview model entry for Gemini 3.1 Pro, suitable for evaluating complex reasoning, long material synthesis, and advanced code collaboration capabilities in a controlled environment. It can be used for problem analysis prototyping, cross-document research assistants, complex plan generation, code reviews, and multi-step tool orchestration. When using it, the source of facts, calculation rules, and tools allowed to be called should be clearly written, and the effect should be tested with task success rate and manual scoring. As it is still in the preview phase, interface behavior, availability and performance may change, requiring alternatives for critical production processes.
- Context
- 1048576 tokens
- Max output
- 65536 tokens
- Input
- llm image video text-to-speech
- Output
- llm
Pricing
Input: $2 / 1M tokens
Output: $12 / 1M tokens
Cached input: $0.2 / 1M tokens
Cache write: $4.5 / 1M tokens
Endpoints
- gemini
- openai