University of California, San Diego

TritonAI Developer API

TritonAI provides programmatic access to a curated selection of large language models through a secure, centralized LLM gateway powered by LiteLLM. It supports chat, reasoning, vision, image generation, OCR, and coding. Access is restricted to approved UC San Diego faculty, staff, researchers, and campus teams, authenticated with issued API keys.

GitHub OpenAPI

Documentation

📖
Documentation
https://tritonai.ucsd.edu/developer-apis/index.html

Specifications

⚙
OpenAPI
https://raw.githubusercontent.com/api-evangelist/ucsd/refs/heads/main/openapi/ucsd-tritonai.yaml

Other Resources

🔗
SignUp
https://tritonai.ucsd.edu/developer-apis/index.html
🔗
x-json-schema
https://raw.githubusercontent.com/api-evangelist/ucsd/refs/heads/main/json-schema/ucsd-chat-completion-request-schema.json
🔗
x-json-schema
https://raw.githubusercontent.com/api-evangelist/ucsd/refs/heads/main/json-schema/ucsd-embedding-request-schema.json
🔗
x-json-structure
https://raw.githubusercontent.com/api-evangelist/ucsd/refs/heads/main/json-structure/ucsd-chat-completion-structure.json
🔗
x-json-structure
https://raw.githubusercontent.com/api-evangelist/ucsd/refs/heads/main/json-structure/ucsd-embedding-structure.json
🔗
x-example
https://raw.githubusercontent.com/api-evangelist/ucsd/refs/heads/main/examples/ucsd-chat-completion-example.json
🔗
x-example
https://raw.githubusercontent.com/api-evangelist/ucsd/refs/heads/main/examples/ucsd-embeddings-example.json
🔗
x-spectral-rules
https://raw.githubusercontent.com/api-evangelist/ucsd/refs/heads/main/rules/ucsd-rules.yml
🔗
x-vocabulary
https://raw.githubusercontent.com/api-evangelist/ucsd/refs/heads/main/vocabulary/ucsd-vocabulary.yml
🔗
x-json-ld
https://raw.githubusercontent.com/api-evangelist/ucsd/refs/heads/main/json-ld/ucsd-context.jsonld

OpenAPI Specification

ucsd-tritonai.yaml Raw ↑ 
openapi: 3.0.3
info:
  title: TritonAI Developer API (LiteLLM Gateway)
  description: >-
    OpenAI-compatible LLM gateway operated by UC San Diego, powered by LiteLLM.
    This OpenAPI 3.0 description was faithfully derived from the live LiteLLM
    OpenAPI 3.1 document published at https://tritonai-api.ucsd.edu/openapi.json
    and scoped to the core, OpenAI-compatible inference surface that TritonAI
    documents (chat, completions, embeddings, model listing, image generation,
    and audio). Access is restricted to approved UC San Diego faculty, staff,
    researchers, and campus teams authenticated with issued API keys.
  version: 1.0.0
  contact:
    name: TritonAI
    email: [email protected]
    url: https://tritonai.ucsd.edu/developer-apis/index.html
servers:
  - url: https://tritonai-api.ucsd.edu
    description: TritonAI LLM gateway
security:
  - bearerAuth: []
tags:
  - name: chat
    description: Chat completion endpoints.
  - name: completions
    description: Text completion endpoints.
  - name: embeddings
    description: Embedding generation endpoints.
  - name: models
    description: Model discovery endpoints.
  - name: images
    description: Image generation endpoints.
  - name: audio
    description: Speech synthesis and transcription endpoints.
paths:
  /v1/chat/completions:
    post:
      tags: [chat]
      summary: Chat Completion
      operationId: chat_completion
      description: >-
        Create a model response for the given chat conversation. Follows the
        OpenAI chat completions request/response contract.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ChatCompletionRequest'
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                type: object
                additionalProperties: true
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
  /v1/completions:
    post:
      tags: [completions]
      summary: Completion
      operationId: completion
      description: Create a text completion for the provided prompt and model.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/CompletionRequest'
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                type: object
                additionalProperties: true
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
  /v1/embeddings:
    post:
      tags: [embeddings]
      summary: Embeddings
      operationId: embeddings
      description: Create an embedding vector representing the input text.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/EmbeddingRequest'
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                type: object
                additionalProperties: true
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
  /v1/models:
    get:
      tags: [models]
      summary: Model List
      operationId: model_list
      description: List the models available through the gateway.
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                type: object
                additionalProperties: true
  /v1/images/generations:
    post:
      tags: [images]
      summary: Image Generation
      operationId: image_generation
      description: Generate one or more images from a text prompt.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              type: object
              required: [model, prompt]
              properties:
                model:
                  type: string
                prompt:
                  type: string
                n:
                  type: integer
                size:
                  type: string
              additionalProperties: true
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                type: object
                additionalProperties: true
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
  /audio/speech:
    post:
      tags: [audio]
      summary: Audio Speech
      operationId: audio_speech
      description: Generate audio (text-to-speech) from input text.
      responses:
        '200':
          description: Successful Response
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
  /audio/transcriptions:
    post:
      tags: [audio]
      summary: Audio Transcriptions
      operationId: audio_transcriptions
      description: Transcribe audio into the input language (speech-to-text).
      responses:
        '200':
          description: Successful Response
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
components:
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: Issued TritonAI API key passed as a Bearer token.
  schemas:
    ChatCompletionRequest:
      type: object
      required: [model, messages]
      properties:
        model:
          type: string
          title: Model
        messages:
          type: array
          items:
            $ref: '#/components/schemas/ChatCompletionMessage'
        temperature:
          type: number
        top_p:
          type: number
        n:
          type: integer
        stream:
          type: boolean
        max_tokens:
          type: integer
      additionalProperties: true
    ChatCompletionMessage:
      type: object
      required: [role]
      properties:
        role:
          type: string
          enum: [system, user, assistant, tool, function, developer]
        content:
          oneOf:
            - type: string
            - type: array
              items:
                $ref: '#/components/schemas/ChatCompletionContentObject'
        name:
          type: string
      additionalProperties: true
    ChatCompletionContentObject:
      type: object
      required: [type]
      properties:
        type:
          type: string
          enum: [text, image_url, input_audio]
        text:
          type: string
        image_url:
          oneOf:
            - type: string
            - type: object
              properties:
                url:
                  type: string
              additionalProperties: true
      additionalProperties: true
    CompletionRequest:
      type: object
      required: [model, prompt]
      properties:
        model:
          type: string
          title: Model
        prompt:
          oneOf:
            - type: string
            - type: array
              items:
                type: string
        max_tokens:
          type: integer
        temperature:
          type: number
        stream:
          type: boolean
      additionalProperties: true
    EmbeddingRequest:
      type: object
      required: [model]
      properties:
        model:
          type: string
          title: Model
        input:
          type: array
          default: []
          items:
            type: string
        timeout:
          type: integer
          default: 600
      additionalProperties: true
    HTTPValidationError:
      type: object
      title: HTTPValidationError
      properties:
        detail:
          type: array
          title: Detail
          items:
            $ref: '#/components/schemas/ValidationError'
    ValidationError:
      type: object
      title: ValidationError
      required: [loc, msg, type]
      properties:
        loc:
          type: array
          title: Location
          items:
            oneOf:
              - type: string
              - type: integer
        msg:
          type: string
          title: Message
        type:
          type: string
          title: Error Type
    ErrorResponse:
      type: object
      title: ErrorResponse
      required: [detail]
      properties:
        detail:
          type: object
          title: Detail
          additionalProperties: true