University of California, San Diego · JSON Structure

Ucsd Chat Completion Structure

JSON Structure for the TritonAI chat completion request object (POST /v1/chat/completions).

Type: object Properties: 7 Required: 2
EducationHigher EducationUniversityResearchUnited StatesCalifornia

TritonAIChatCompletionRequest is a JSON Structure definition published by University of California, San Diego, describing 7 properties, of which 2 are required. It conforms to the https://json-structure.org/meta/core/v0 meta-schema.

Properties

model messages temperature top_p n stream max_tokens

Meta-schema: https://json-structure.org/meta/core/v0

JSON Structure

Raw ↑ 
{
  "$schema": "https://json-structure.org/meta/core/v0",
  "$id": "https://raw.githubusercontent.com/api-evangelist/ucsd/main/json-structure/ucsd-chat-completion-structure.json",
  "name": "TritonAIChatCompletionRequest",
  "description": "JSON Structure for the TritonAI chat completion request object (POST /v1/chat/completions).",
  "type": "object",
  "properties": {
    "model": { "type": "string", "description": "Model identifier." },
    "messages": {
      "type": "array",
      "description": "Ordered conversation messages.",
      "items": {
        "type": "object",
        "name": "ChatMessage",
        "properties": {
          "role": { "type": "string", "description": "Message role: system, user, assistant, tool, function, or developer." },
          "content": { "type": "string", "description": "Message content; may be plain text or an array of typed content parts." },
          "name": { "type": "string", "description": "Optional participant name." }
        },
        "required": ["role"]
      }
    },
    "temperature": { "type": "number", "description": "Sampling temperature." },
    "top_p": { "type": "number", "description": "Nucleus sampling probability mass." },
    "n": { "type": "int32", "description": "Number of completions to generate." },
    "stream": { "type": "boolean", "description": "Whether to stream partial results." },
    "max_tokens": { "type": "int32", "description": "Maximum tokens to generate." }
  },
  "required": ["model", "messages"]
}