# Gemini 3.1 Flash Live

> Audio-to-audio realtime dialogue on the Gemini 3 line, at the same per-minute price as Gemini 2.5 Flash Native Audio. The successor to that model, which retires 2026-10-20. WebSocket-based; native audio in and out in one round-trip.

Human version: https://kymaapi.com/models/gemini-3.1-flash-live-preview
Live JSON: `GET https://kymaapi.com/v1/models` (no auth required)

## Facts

- **Model ID**: `gemini-3.1-flash-live-preview` — pass this as `model` in the request body
- **Creator**: Google
- **Released**: 2026-03-26
- **Price**: $0.03888 / minute

## Call it

```bash
curl https://kymaapi.com/v1/audio/speech \
  -H "Authorization: Bearer $KYMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "gemini-3.1-flash-live-preview", "prompt": "..."}'
```

## Positioning

A WebSocket-based audio-to-audio model designed for low-latency spoken dialogue and voice agents. Reach for it when migrating from Gemini 2.5 Flash Native Audio or when you need native round-trip audio processing.

## About

Gemini 3.1 Flash Live is a realtime audio-to-audio model from Google. It processes native audio input and returns native audio output in a single WebSocket round-trip, supporting a 131,072-token context window and up to 65,536 output tokens.

On Kyma, it runs behind an OpenAI-compatible endpoint with automatic request failover if a serving path degrades. New accounts receive $0.50 in free credit, and prompt caching is supported, billing repeated prefixes at 10% of the standard input rate. The platform returns exact request cost in usage.cost and the active model ID in the X-Kyma-Model header.

The model does not support reasoning, vision, or structured outputs. It operates strictly on audio modalities and is billed per minute. It serves as the direct successor to Gemini 2.5 Flash Native Audio, which retires on 2026-10-20.

## Use cases

- **Low-Latency Voice Agents** — Handles real-time spoken conversations with native audio input and output in a single WebSocket connection.
- **Speech Generation Workflows** — Converts audio prompts into synthesized speech without intermediate transcription steps.
- **Interactive Spoken Chat** — Powers conversational interfaces that require immediate audio responses for hands-free applications.

## Not ideal for

Do not use this model if your application requires text-based reasoning, image understanding, or strictly formatted JSON outputs.

## Pick something else when

- You need live translation capabilities → `gemini-3.5-live-translate-preview`
- You need multilingual real-time translation → `gpt-realtime-translate`
- You need to maintain the older 2.5 architecture → `gemini-2.5-flash-native-audio-preview-12-2025`

## See also

- All models: https://kymaapi.com/models.md
- Pricing: https://kymaapi.com/pricing.md
- Other models by Google: https://kymaapi.com/models?q=Google