Gemini 3 Pro Preview
RetiredA Google model with a 1M-token context window, image input and tool calling. Released on the Gemini API November 18, 2025, now retired.
Access details: Gemini API
The model ID to use in API requests.
gemini-3-pro-previewUpdated Sep 23, 2026
Capabilities & limits
- Context window
- 1M
- Maximum output
- 64K
- Image input
- Supported
- Tool calling
- Supported
- Structured output
- Supported
Pricing
USD / 1M tokens- Input
- No longer sold
- Output
- No longer sold
No longer sold on the Gemini API.
Price history
Mar 9, 2026
Input No longer sold per million tokens
Mar 9, 2026
Output No longer sold per million tokens
Technical details
- Parameters
- Not published
- Active parameters
- Not published
- Architecture
- Sparse mixture-of-experts (MoE) transformer with native multimodal support for text, vision, and audio inputs
- Training / knowledge cutoff
- January 2025
- Input modalities
- Text, image, video, audio, PDF
- Output modalities
- Text
From Google
Official announcement ↗Release & lifecycle
- First available
- Nov 18, 2025
- retired
- Mar 9, 2026