Hugging Face · Schema

ChatCompletionRequest

Properties

Name Type Description
model string Model ID to use. Can be a Hugging Face model ID (e.g., meta-llama/Llama-3-70b-chat-hf) or a provider-specific identifier.
messages array List of messages comprising the conversation
frequency_penalty number Penalize tokens based on frequency in text so far
logprobs boolean Whether to return log probabilities
max_tokens integer Maximum number of tokens to generate
presence_penalty number Penalize tokens based on presence in text so far
reasoning_effort string Constrains effort on reasoning for models that support it. Common values are none, minimal, low, medium, high, xhigh.
response_format string
seed integer Random seed for reproducibility
stop array Up to 4 sequences where generation will stop
stream boolean Whether to stream partial responses using SSE
stream_options object
temperature number Sampling temperature
tool_choice string Controls tool usage
tool_prompt string Prompt prepended before tools
tools array List of tools the model may call
top_logprobs integer Number of most likely tokens to return per position
top_p number Nucleus sampling parameter
View JSON Schema on GitHub