Developer APIs & Resources

Overview

The TritonAI Developer API provides UCSD faculty, staff, and researchers with programmatic access to a curated selection of large language models (LLMs) through a unified API gateway powered by LiteLLM. This service enables you to integrate AI capabilities directly into your applications, research workflows, and automation projects while maintaining compliance with UC data protection requirements.

Whether you're building intelligent document processing pipelines, automating data analysis, enhancing research workflows, or prototyping innovative AI applications, the Developer API gives you the tools to bring your ideas to life.

Key Features

• Unified API Access: Connect to multiple AI models through a single, standardized API endpoint. Our LiteLLM gateway abstracts away the complexity of working with different model providers, giving you a consistent interface regardless of which model you choose.
• Enterprise Security & Compliance: All API traffic is routed through UCSD's secure infrastructure. Cloud provider models operate under campus-wide enterprise agreements with Azure, GCP, and AWS. Self-hosted models keep your data entirely within UC San Diego's walls. No data is used for model training.
• Flexible Model Selection: Choose from a variety of models optimized for different use cases — from fast, cost-effective options for simple tasks to advanced models for complex reasoning and analysis. Select cloud-hosted models for cutting-edge capabilities or self-hosted models for maximum data control.
• Transparent Pricing: Clear, predictable pricing based on token usage with no hidden fees. Pay only for what you use beyond your initial free credits.
• Usage Tracking & Reporting: Monitor your API consumption in real-time and access detailed usage reports for project accounting and budgeting.

Model Hosting & Data Protection

The TritonAI Developer API provides access to models through two distinct hosting environments, giving you flexibility to choose the right option based on your data sensitivity requirements.

Cloud Provider Models

We offer access to leading commercial models from major cloud providers, including Microsoft Azure, Google Cloud Platform (GCP), and Amazon Web Services (AWS), all operating under UC San Diego's campus-wide enterprise agreements.

These agreements provide enterprise data protection terms that meet UC system requirements, ensure your data is not used for model training, and include contractual safeguards that individual accounts cannot obtain.

Self-Hosted Models

For projects requiring the highest level of data control, we offer locally-hosted open-source and open-weight models running entirely within UC San Diego infrastructure at the San Diego Supercomputer Center.

With self-hosted models, all inference processing occurs on-premises, so your prompts and responses never leave campus infrastructure. This makes them ideal for research involving sensitive data, projects with heightened privacy considerations, or any use case where keeping data on-premises is preferred or required.

Our model list indicates the hosting environment for each available model, making it easy to select options that meet your requirements.

Available Models & Pricing

We offer a curated selection of large language models to meet diverse needs and budgets. Our model catalog is regularly updated as the AI landscape evolves, ensuring you have access to the latest capabilities.

View Complete Model List & Pricing

Our model catalog includes options across multiple providers and capabilities:

• Chat & reasoning models from Anthropic, OpenAI, Google, Mistral, and others
• Vision-enabled models for image understanding and analysis
• Image generation models for creating visual content
• Specialized models for coding, OCR, and advanced reasoning tasks
• Self-hosted models for maximum data control

The model hub displays current per-token pricing (input and output rates), context window sizes, available features, and hosting environment for each model. Pricing is based on token usage — you pay only for what you consume.

Technical Details

API Compatibility

Our LiteLLM gateway provides an OpenAI-compatible API interface, making it easy to integrate with existing tools, libraries, and codebases. If your application works with the OpenAI API, it will work with Triton AI with minimal modifications.

Authentication

API access is secured via API keys issued upon approval. Keys should be stored securely and never committed to version control or shared publicly.

Rate Limits

Default rate limits are designed to support typical development and production workloads. Projects requiring higher throughput can request limit increases.

Documentation & Support

Approved users receive access to:

• Complete API documentation
• Code examples in Python, JavaScript, and other languages
• Integration guides for common frameworks
• Support resources and community channels

Billing & Recharge Rates

Free Credits

All new users receive $50 in free credits upon approval. Credits do not expire but are non-transferable.

Recharge Rates

Usage beyond free credits is billed monthly to your designated chart string at the rates published in our model pricing list. Recharge rates reflect the actual cost of model access plus a necessary infrastructure fee to sustain the service.

Project & Task Tracking

When requesting access, you'll provide project and task information that maps to your chart string. This enables accurate cost allocation and simplifies reporting for grant-funded or departmentally-funded projects.

Usage Alerts

Set up usage alerts to notify you when spending reaches defined thresholds, helping you stay within budget and avoid surprises.