API Endpoints

API Endpoints

Text Generation

• Endpoint: POST /v1/chat/completions
• Parameters:
  • model (string): Model identifier (e.g., “qwq-32b”).
  • messages (array): Array of message objects with:
    • role (string): “system”, “user”, or “assistant”
    • content (string): Message content
  • max_tokens (int): Max tokens to generate.
  • temperature (float): Controls randomness (0.0-1.0).

Speech Recognition

• Endpoint: POST /v1/audio/transcriptions
• Parameters:
  • file (file): Audio file (multipart/form-data).
  • model (string): Model identifier (e.g., “whisper-large-v3”).
  • language (string, optional): Language code.
  • response_format (string, optional): Output format (“json”, “text”, “srt”, “vtt”).

Speech Translation

• Endpoint: POST /v1/audio/translations
• Parameters:
  • file (file): Audio file (multipart/form-data).
  • model (string): Model identifier (e.g., “whisper-large-v3”).
  • response_format (string, optional): Output format (“json”, “text”, “srt”, “vtt”).

Image Generation

• Endpoint: POST /v1/images/generations
• Parameters:
  • model (string): Model identifier (e.g., “flux-realistic”, “flux-illustrate”).
  • prompt (string): Description of the desired image.
  • n (integer, optional): Number of images to generate.
  • size (string, optional): Image dimensions (e.g., “1024x1024”).
  • style (string, optional): For flux-illustrate, artistic style.
  • response_format (string, optional): Output format (“url” or “b64_json”).

[See model-specific docs for more endpoints.]