VETRIN API v2.4

Quickstart

The Vetrin API provides access to medical-grade veterinary intelligence. Integrate specialized animal health diagnostics into your application in minutes.

AI Core

Powered by our specialized medical engine.

Safe & Secure

Built-in medical safety and privacy rules.

Live Speed

Fast responses for busy companies.

Grows with You

Built for companies of any size.

API Keys

Your secret key is how we know it's you. Keep it safe and don't share it.

Authorization: Bearer vt_key_••••••••

Never share your API key or expose it in client-side code. Use environment variables for secure storage.

Chat with AI

Talk to our AI to get medical help and advice. It uses the same rules as popular AI models.

POST/v1/chat/completions

Request Body

ParameterTypeDescription
modelRequired
string

ID of the model to use. Supporting 'vetx-23' and 'vetx-23-vision'.

messagesRequired
array

A list of messages comprising the conversation history. Content can be a string or an array of objects (text/image_url).

temperature
number

Controls randomness. Default is 0.7.

stream
boolean

If set, partial message deltas will be sent via SSE.

max_tokens
integer

The maximum number of tokens to generate.

Vision Support

The vetx-23 engine supports image analysis for clinical diagnostics. Send medical scans, photos of symptoms, or lab results for instant analysis.

How it works

Instead of a string for message content, pass an array containing a text object and an image_url object. Images should be provided as base64-encoded Data URLs.

"content": [
  { "type": "text", "text": "Analyze this..." },
  { "type": "image_url", "image_url": { "url": "data:image/jpeg;base64,..." } }
]

Live Responses

Get answers character by character as they are made. This makes the AI feel faster and more alive.

Implementation Details

When stream: true is passed, the server maintains an open connection. Data chunks are prefixed with data: . When the generation finishes, the server sends a [DONE] message.

Real-time UX

Update your UI character by character for a fluid "thinking" feel.

Early Abort

Terminate requests early to save tokens if the AI veers off-track.

AI Models

Special AI made just for veterinary companies and animal care.

vetx-23

Surgical Precision Engine
Vision + Diagnostic

Our flagship model, optimized for complex diagnostics, pharmacology calculations, and cross-species triage. Supporting multimodal vision analysis via 'vetx-23-vision'.

Context128k
Input$8.32 / 1M
Output$8.32 / 1M

Problems

Standard codes to tell you if things went well or if there was a problem.

CodeDescription
400Bad Request

The request was unacceptable, often due to missing parameters.

401Unauthorized

No valid API key provided.

402Insufficient Balance

Your wallet does not have enough tokens to fulfill the request.

429Rate Limit Reached

Too many requests hit the API too quickly.

500Server Error

Something went wrong on our end.

Usage & Costs

Check how much you've used the AI and what it costs.

GET/v1/usage

Query Parameters

ParameterTypeDescription
from
string

Start date in ISO format (e.g. 2024-01-01T00:00:00Z).

to
string

End date in ISO format.

Speed Limits

How many requests you can make in a minute.

Tier 1 (Free)

20 requests / min

Tier 2 (Pro)

500 requests / min

If you hit these limits, you'll receive a 429 error.

VETRIN AI NETWORK

v2.4.0 • Surgical Intelligence • © 2026