Skip to main content
Each model’s input and output prices, per 1M tokens, are listed on the Models page in AI Gateway (Manage Models).

List of supported models

Responses API

Supported Models

Chat models

Completion models

Embedding models

Image models

Image Generation

Image Edit

Image Variations

Supported Image Models

Moderations models

Classify models

Chat models in this table answer through structured outputs and report their own probabilities; typesafe/jev-latest is the native classify model. See Classify for the question types, and the Create Classify reference for the full request and response schema.

Rerank models

OCR models

Speech-to-Text models

Text-to-Speech models

Text-to-Speech Voices

The following voices are available for Text-to-Speech models:

OpenAI

  • alloy: Neutral, versatile voice
  • echo: Neutral, soft-spoken voice
  • fable: Expressive, narrative-focused voice
  • onyx: Deep, authoritative voice
  • nova: Warm, natural voice
  • shimmer: Clear, optimistic voice

ElevenLabs

  • aria: Neutral, versatile voice
  • roger: Deep, authoritative voice
  • sarah: Warm, friendly voice
  • laura: Soft, gentle voice
  • charlie: Casual, conversational voice
  • george: Professional, articulate voice
  • callum: Youthful, energetic voice
  • river: Calm, soothing voice
  • liam: Clear, confident voice
  • charlotte: Elegant, refined voice
  • alice: Bright, cheerful voice
  • matilda: Thoughtful, measured voice
  • will: Reliable, trustworthy voice
  • jessica: Engaging, expressive voice
  • eric: Authoritative, commanding voice
  • chris: Friendly, approachable voice
  • brian: Mature, distinguished voice
  • daniel: Versatile, balanced voice
  • lily: Sweet, melodious voice
  • bill: Grounded, authentic voice

Retries & Error Handling Streaming