API docs
Use the gateway from your code.
The Utah AI Gateway API works like the OpenAI API. Any OpenAI library works: change the base URL and use your gateway key.
1. Get a key
Create an account, request API access in your portal, then create a key. Students with a .edu email are approved instantly.
2. Make a request
Base URL: https://utahaigateway.com/v1
curl https://utahaigateway.com/v1/chat/completions \
-H "Authorization: Bearer $UTAH_AI_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "utah-fast", "messages": [{"role": "user", "content": "Explain photosynthesis in two sentences."}]}'
Python
from openai import OpenAI
client = OpenAI(base_url="https://utahaigateway.com/v1", api_key="uag_...")
reply = client.chat.completions.create(
model="utah-fast",
messages=[{"role": "user", "content": "Write a haiku about Bryce Canyon."}],
)
print(reply.choices[0].message.content)
Models
| Model name | Best for |
|---|---|
| utah-fast | Fast and inexpensive. Good for chat, summaries and classroom tools. |
| utah-smart | Stronger reasoning and writing. Good for coding help and research. |
Limits and safety
- Each account has a requests-per-minute limit and a daily token limit. You see both in your portal. Over a limit, the API answers
429. - Every request is screened by a safety check. Requests that break the safety policy are blocked with
400and reviewed by an administrator. - Optional: send
X-Gateway-Project: <project id>to count usage toward one of your showcase projects. - Streaming is not supported yet. Send
"stream": false(the default).