Gemini 3.8 Live

As of 2026-11-12, this model is not available for live sessions through APIMaster. The upstream WebSocket handshake returns 403; native Live forwarding and billing have not passed acceptance checks. Registered prices do not establish availability.

Use cases

Accepts text, audio, images and video; outputs text and audio. Suitable for voice support, language practice and live assistants with visual context. Live handles conversations; use TTS for narration and exact script reading.

Official prices

Google Standard list prices, checked on 2026-11-12. These are not current APIMaster sale prices.

Modality	USD / million tokens
Text input	$0.75
Text output (including thinking)	$4.50
Audio input	$3.00
Audio output	$12.00
Image / video input	$1.00

Google direct reference example

This example connects directly to Google and requires a Google API key with Live access, not an APIMaster key. It does not imply that an APIMaster Live endpoint is currently available.

Install the SDK, set GEMINI_API_KEY in your server environment, then run the program. It saves a short spoken reply to reply.wav.

python -m pip install google-genai
python gemini_live_reference.py

import asyncio
import os
import wave
from google import genai

async def main():
    client = genai.Client(api_key=os.environ["GEMINI_API_KEY"])
    async with client.aio.live.connect(
        model="gemini-3.8-live",
        config={"response_modalities": ["AUDIO"]},
    ) as session:
        await session.send_client_content(
            turns={"role": "user", "parts": [{"text": "Say hello briefly."}]},
            turn_complete=True,
        )
        audio = bytearray()
        async for message in session.receive():
            content = message.server_content
            if content and content.model_turn:
                for part in content.model_turn.parts or []:
                    if part.inline_data and part.inline_data.data:
                        audio.extend(part.inline_data.data)
        if not audio:
            raise RuntimeError("The Live session returned no audio")
        with wave.open("reply.wav", "wb") as output:
            output.setnchannels(1)
            output.setsampwidth(2)
            output.setframerate(24000)
            output.writeframes(audio)

asyncio.run(main())

Audio input and output

Microphone input uses 16 kHz mono signed 16-bit little-endian PCM; output uses 24 kHz PCM. Base64-encode audio and send chunks with the following message; preserve protocol field names.

{"realtimeInput":{"audio":{"mimeType":"audio/pcm;rate=16000","data":"BASE64_PCM16_AUDIO"}}}

Sessions and troubleshooting

Use the Live API WebSocket. Wait for setupComplete before sending input; turnComplete ends a turn. A handshake 403 means no connection was established, not a fingerprint mismatch. generateContent and Chat Completions do not replace a bidirectional Live session. Production acceptance must cover authentication, audio, interruption, tools, reconnects and modality-aware billing. Keep keys on the server, outside public pages and repositories.

Sources and verification date

2026-11-12 · Google: Gemini 3.8 Live · Google: Live API · Google: pricing