Gemini 3.8 Flash TTS

Third-party

Gemini 3.8 Flash TTS is Google's flagship text-to-speech model for expressive, natural speech generation.

Model Info
Context Window ↗	8,192 tokens
More information	link ↗
Pricing	Input text (per 1M tokens)$0.50 Cached input (per 1M tokens)$0.125 Output audio (per 1M tokens)$9.00

Usage

const response = await env.AI.run(
  'google/gemini-3.8-flash-tts',
  { text: 'Hello, welcome to Cloudflare AI Gateway!' },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "google/gemini-3.8-flash-tts",
  "input": {
    "text": "Hello, welcome to Cloudflare AI Gateway!"
  }
}'

{
  "audio": "https://examples.aig.cloudflare.com/google/gemini-3.8-flash-tts/simple-text-to-speech.wav",
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

Examples

Expressive Pause — Generate expressive speech with an inline vocal pause

const response = await env.AI.run(
  'google/gemini-3.8-flash-tts',
  {
    text: 'Welcome to the future of voice. <short pause> Let us build something remarkable together.',
    voice: 'Puck',
  },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "google/gemini-3.8-flash-tts",
  "input": {
    "text": "Welcome to the future of voice. <short pause> Let us build something remarkable together.",
    "voice": "Puck"
  }
}'

{
  "audio": "https://examples.aig.cloudflare.com/google/gemini-3.8-flash-tts/expressive-pause.wav",
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

Narrated Passage — Generate a longer narrated passage with a different voice

const response = await env.AI.run(
  'google/gemini-3.8-flash-tts',
  {
    text: 'At sunrise, the research team opened the observatory and watched the first light move across the valley. Every sensor came online in sequence, and the quiet room filled with the soft rhythm of discovery.',
    voice: 'Charon',
  },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "google/gemini-3.8-flash-tts",
  "input": {
    "text": "At sunrise, the research team opened the observatory and watched the first light move across the valley. Every sensor came online in sequence, and the quiet room filled with the soft rhythm of discovery.",
    "voice": "Charon"
  }
}'

{
  "audio": "https://examples.aig.cloudflare.com/google/gemini-3.8-flash-tts/narrated-passage.wav",
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

French Greeting — Generate speech from a French greeting using a distinct voice

const response = await env.AI.run(
  'google/gemini-3.8-flash-tts',
  {
    text: "Bonjour et bienvenue. Nous sommes ravis de vous accompagner aujourd'hui.",
    voice: 'Kore',
  },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "google/gemini-3.8-flash-tts",
  "input": {
    "text": "Bonjour et bienvenue. Nous sommes ravis de vous accompagner aujourd'\''hui.",
    "voice": "Kore"
  }
}'

{
  "audio": "https://examples.aig.cloudflare.com/google/gemini-3.8-flash-tts/french-greeting.wav",
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

Parameters

text

stringrequiredmaxLength: 10000The text to convert to speech. Maximum 10,000 characters.

voice

stringenum: Zephyr, Puck, Charon, Kore, Fenrir, Leda, Orus, Aoede, Callirrhoe, Autonoe, Enceladus, Iapetus, Umbriel, Algieba, Despina, Erinome, Algenib, Rasalgethi, Laomedeia, Achernar, Alnilam, Schedar, Gacrux, Pulcherrima, Achird, Zubenelgenubi, Vindemiatrix, Sadachbia, Sadaltager, SulafatThe voice to use for speech synthesis

temperature

numberminimum: 0maximum: 2Controls randomness in generation (0-2)

topP

numberminimum: 0maximum: 1Nucleus sampling threshold (0-1). Tokens with cumulative probability up to topP are considered

topK

integerexclusiveMinimum: 0maximum: 9007199254740991Only sample from the top K tokens. Smaller K = more focused, larger K = more diverse

maxOutputTokens

integerexclusiveMinimum: 0maximum: 9007199254740991Maximum number of tokens to generate

▶stopSequences[]

arraySequences where the model will stop generating further tokens

audio

stringBase64-encoded audio data (WAV format)

API Schemas (Raw)

Input

Output