Gemini 3.5 Live Translate is our low-latency, audio-to-audio model optimized for real-time translation of spoken conversations. It enables seamless, bidirectional translation with high accuracy and natural voice output. Try in Google AI Studio
Documentation
Visit the Live Translation guide for full coverage of features and capabilities.
gemini-3.5-live-translate-preview
| Property | Description |
|---|---|
| Model code | gemini-3.5-live-translate-preview |
| Supported data types | Inputs Audio (speech) Output Audio (translated speech) and Text (transcript) |
| Token limits[1](https://ai.google.dev/gemini-api/docs/tokens)^ | Input token limit 131,072 Output token limit 65,536 |
| Capabilities | Audio generation Supported Caching Not supported Code execution Not supported File search Not supported Function calling Not supported Grounding with Google Maps Not supported Image generation Not supported Live API Supported Live translation Supported Search grounding Not supported Structured outputs Not supported Thinking Not supported URL context Not supported |
| Consumption options | Batch API Not supported Flex inference Not supported Priority inference Not supported |
| Versions | Read the model version patterns for more details. - Preview: gemini-3.5-live-translate-preview |
| Latest update | June 2026 |
| Model card | Model card |
[*] ↩︎