• Original Model Card
• text-generation Task on Hugging Face

Send Request

curl <AZUREML_ENDPOINT_URL> \
    -X POST \
    -H "Authorization: Bearer <AZUREML_TOKEN>" \
    -H "Content-Type: application/json" \
    -d '{"inputs":"Write a short story about a robot learning to paint"}'

Supported Parameters

• inputs (string): The input text prompt to generate text from
• parameters (object):
  • do_sample (boolean): Whether to use sampling. Set to false for deterministic output.
  • max_new_tokens (integer): Maximum number of tokens to generate in the output.
  • repetition_penalty (float): Penalty for repeating tokens from the input or previous output.
  • return_full_text (boolean): Whether to return the full text including the prompt.
  • seed (integer): Seed for the random number generator to ensure reproducible results.
  • temperature (float): Controls randomness in generation. Lower values make output more deterministic.
  • top_k (integer): Number of highest probability vocabulary tokens to keep for top-k-filtering.
  • top_p (float): Nucleus sampling parameter. Keeps the smallest set of tokens whose cumulative probability exceeds top_p.
  • truncate (integer): Truncate input to this many tokens.
  • typical_p (float): Typical sampling parameter for locally typical sampling.

OpenAI Chat Completions API