Triton AI Developer API
Build with approved models through the Triton AI Developer API.
Access approved enterprise-cloud and UC-hosted models through one API. Use an OpenAI-compatible client with the Triton AI base URL and your API key.
Your API key determines which models you can use, your limits, and how usage is billed. Browse the public Model Hub to compare models.
Start
Quickstart
Send a request with Python or cURL.
API reference
Find request routes, authentication headers, and response headers.
Models
Compare models by task, limits, and price.
Build
Chat completions
Send text messages and stream a response.
Responses
Use the Responses API for current OpenAI models and coding agents.
Tool calling
Connect a model to functions in your application.
Embeddings
Create vectors with the hosted Qwen Embedding 4B route.
Image input
Send an image to a vision-capable chat model.
Speech to text
Transcribe a recording with Python or cURL.
Privacy Filter
Classify sensitive text with the hosted token-classification model.
Safety and security
Apply data controls and classify requests against a policy.
Deploy
Errors and retries
Handle request failures and rate limits.
Production checklist
Prepare keys, data controls, monitoring, and support.
Integrations
Connect Claude Code, Codex, and other developer tools.
API endpoints
| Resource | URL |
|---|---|
| Base URL | https://tritonai-api.ucsd.edu/v1 |
| Documented routes | Open the API reference |
| Public model data | Open the JSON endpoint |
| Model Hub interface | Open the Model Hub |
Access and support
Submit the Developer API access request to request a key. Access, limits, and available models depend on the approved use case.
Email tritonai@ucsd.edu for support.
Protect your key
Keep the API key out of prompts, messages, screenshots, source files, repositories, and shared configuration.