Gemini 3.8 Flash TTS

# Set OPERATION_ID once per generation; reuse it when retrying.
OPERATION_ID=$(python3 -c "import uuid; print(uuid.uuid4())")
curl https://api.tokely.me/v1/media/tasks \
  -H "Authorization: Bearer $TOKELY_API_KEY" \
  -H "Idempotency-Key: $OPERATION_ID" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "gemini-3.8-flash-tts",
  "prompt": "A quiet coastal village at sunrise.",
  "metadata": {
    "voice": "Kore"
  }
}'

# Read the returned task_id, then check its status:
curl https://api.tokely.me/v1/media/tasks/TASK_ID \
  -H "Authorization: Bearer $TOKELY_API_KEY"

# When complete, list downloadable results:
curl https://api.tokely.me/v1/media/tasks/TASK_ID/artifacts \
  -H "Authorization: Bearer $TOKELY_API_KEY"

Set TOKELY_API_KEY to your dashboard key. Requests require a funded Tokely balance. Base URL: https://api.tokely.me/v1.

Capabilities & limits

Provider

Google

API model ID

gemini-3.8-flash-tts

Category

Audio

Endpoints

POST /v1/media/tasks

Output

Audio · background task

Supported input

Prompt, Voice

Input text

$0.158826 per 1M input tokens

Audio output

$2.858868 per 1M output tokens

API reference

Send duration at the top level, reference images as images, and other model settings in metadata. The Playground generates a matching API request.

Authenticate with Authorization: Bearer YOUR_TOKELY_API_KEY. Use your Tokely base URL and API key.

model

Required string: gemini-3.8-flash-tts

prompt (required)

Text, up to 80 characters

voice

Kore, Zephyr, Puck. Default: Kore

Requests are charged to your Tokely balance using the options and rates shown above. Your dashboard records usage.

Save the returned task ID. Check its status, then list its artifacts and download them with your API key. Stopping the browser’s progress checks does not cancel an accepted generation.

400 / 415

Invalid request, parameter, endpoint for this model, or content type.

401 / 403

Invalid, revoked or restricted API key.

402

Insufficient account balance or API key budget.

429

Request limit reached. Retry after the indicated delay.

502 / 503 / 504

Upstream failure, temporary service unavailability or request timeout.

Compare equivalent generation settings

Match model version, duration, resolution and input requirements before comparing two prices. Results are kept for 30 days; download those you need to keep.

Compare audio API costs · Google models · Methodology