Sign in

Generate a chat message

Generate the next chat message in a conversation between a user and an assistant.

post/api/chat

Request body

modelstring required

Model name

streamboolean
logprobsboolean

Whether to return log probabilities of the output tokens

top_logprobsinteger

Number of most likely tokens to return at each token position when logprobs are enabled

Response

Chat response

modelstring

Model name used to generate this message

created_atstring date-time

Timestamp of response creation (ISO 8601)

doneboolean

Indicates whether the chat response has finished

done_reasonstring

Reason the response finished

total_durationinteger

Total time spent generating in nanoseconds

load_durationinteger

Time spent loading the model in nanoseconds

prompt_eval_countinteger

Number of tokens in the prompt

prompt_eval_durationinteger

Time spent evaluating the prompt in nanoseconds

eval_countinteger

Number of tokens generated in the response

eval_durationinteger

Time spent generating tokens in nanoseconds

Changes

Changed in 3 of the 9 revisions of this API.5

  • d54e2ef5c24a1See the full diff
    • ○

      added the new max enum value to the request property think/oneOf[subschema #2]/

      request-property-enum-value-added

  • 3e72f42e8aa63See the full diff
    • ○

      added the new optional request property logprobs

      new-optional-request-property

    • ○

      added the new optional request property top_logprobs

      new-optional-request-property

    • ○

      added the optional property logprobs to the response with the 200 status

      response-optional-property-added

  • c99d500c9ea41See the full diff
    • ○

      request property think list-of-types was widened by adding types string to media type application/json

      request-property-list-of-types-widened