Google: Gemini 3.5 Live Translate
Gemini 3.5 Live Translate is a real-time speech translation model in the Gemini API. Google's models page describes it as a "low-latency, real-time speech to speech translation model that supports 70+ languages." The model id is gemini-3.5-live-translate-preview, and the page lists it in the Preview section, tagged New.
Status
Preview. The model id carries the -preview suffix and sits under Google's Preview heading. Google's deprecations page lists it with no shutdown date announced and no replacement. Check the docs before depending on it in production.
Key Features
- Speech to speech: it translates spoken input into spoken output rather than going through text only.
- 70+ languages: language coverage as stated on Google's models page.
- Low latency: designed for live conversation, per Google's description.
Google's models page does not state input and output formats or token limits for this model, so this page does not either.
Use Cases
- Live interpretation in meetings, support calls and events.
- Multilingual voice experiences where waiting for a text transcript is too slow.
- Prototyping translated voice agents before a stable model is released.
Limitation: it is a preview model, so interfaces, quotas and behavior may change.
Pricing
Not verified from Google's pricing page on the date checked, so no price is quoted here. See Google's Gemini API pricing page for current figures.
Getting Started
- Open the Gemini models page and find the Live Translate entry for the exact id.
- Request
gemini-3.5-live-translate-previewthrough the Gemini API with a Google AI Studio key. - Test your target language pairs before shipping, since the page does not publish per-language quality figures.
Frequently Asked Questions
What is Gemini 3.5 Live Translate?
A real-time speech-to-speech translation model in the Gemini API, supporting 70+ languages.
Is it generally available?
Not yet. Google lists it as a preview model with the id gemini-3.5-live-translate-preview.
Does it have a shutdown date?
Google's deprecations page lists no shutdown date and no replacement for it.
Is it a general chat model?
No. Google describes it as a speech translation model. For general chat, see Google's Flash models.
Alternatives
- Gemini 3.1 Flash Live: another live-audio Gemini model in this catalog.
- Gemini 3.6 Flash: a stable Flash model for general tasks.
- Gemini 3.5 Flash: a general-purpose Gemini Flash model.
Conclusion
Gemini 3.5 Live Translate targets real-time speech-to-speech translation across 70+ languages. It is still in preview, so test carefully and follow the Gemini models page for status changes.
Sources
- Websitehttps://ai.google.dev/gemini-api/docs/models
- Social Mediahttps://x.com/GoogleAI
Comments
No comments yet. Be the first to comment!
Related Tools
Related Insights

After Code Stopped Being Scarce (Part II): From AI Writing Code to AI Taking Part in Delivery
A year after leaving my previous employer, I document the delivery workflow for our financial business that we built around Claude Code and Opus 5.5: XaC, progressive rules, independent review and testing by Grok Bot, PR watch, and our actual delivery over the past three months.
Point Codex at any model with magpie
Install magpie, add a local or signed-in provider, then switch Codex to that model. Rechecked against official magpie docs on 2026-10-06.