Introduction

A single, OpenAI-compatible endpoint for healthcare AI - symptom checking, wellness management, emergency assist, patient journey support, and more.

What is Vaidya AI API?

Vaidya AI API is an OpenAI/vLLM-compatible Chat Completions endpoint built for healthcare. Send a standard model and messages body — the model implicitly understands your healthcare request and routes to the right capability.

Why use it?

One endpoint, many use cases - Symptom Checker, Wellness Management, Emergency Assist, Patient Journey Assist, Administrator Assist, and more behind a single POST.
Drop-in compatible - works with the OpenAI Python/JS SDK, requests, curl, or any HTTP client.
Predictable pricing - credit-based model so you know the cost per call before you ship.

Vaidya Model 2.0

Vaidya 2.0 is powered by our latest innovations in LLM post-training techniques, a deep-research augmented reasoning framework, and doctor-annotated synthetic healthcare data creation. Our rubric-based reinforcement learning trains Vaidya to think across the full spectrum of what makes a great clinical response - accuracy, completeness, empathy, and context. Parallel agentic thinking supercharges Vaidya 2.0 with the ability to deep-search the web, gathering and reasoning on top of the latest and most reliable healthcare information.

Benchmark performance

Vaidya 2.0 was rigorously evaluated on OpenAI's HealthBench Hard and outperformed every major model globally - including GPT-5 Thinking, Gemini-3-Pro, and Baichuan-M3-235B.

Rank	Model	HealthBench Hard Score
1	Vaidya 2.0 by Fractal	50.1
2	OpenAI GPT-5 Thinking	46.2
3	Baichuan-M3-235B	44.6
4	OpenAI GPT-5.2 Thinking	42.1
5	OpenAI GPT-5 Mini	40.3
6	Google Gemini 3 Pro	25.5
7	Qwen3 A3B-30B	18.7
8	Google MedGemma 27B	15.5

Supported Use Cases

Vaidya AI API supports a range of healthcare use cases. The model implicitly understands your request and responds accordingly — no special parameter required:

Use Case	What it does
Symptom Checker	Triage symptoms and provide safe next steps and red flags
Wellness Management	Personalized health scoring, plans, and lifestyle recommendations
Emergency Assist	Rapid guidance for urgent health situations
Patient Journey Assist	End-to-end support across care milestones and follow-ups
Administrator Assist	Support for clinical and administrative workflows

Requests are priced on a credit-based model. See Monitoring Usage for quota and billing details.

Quick overview


Base URL	`https://api.vaidya.ai`
Endpoint	`POST vaidya/chat/completions`
Auth	Bearer token - get your key from console.vaidya.ai
Format	Standard OpenAI chat request

Minimal request

The smallest valid call matches OpenAI Chat Completions:

curl --request POST \
  --url https://api.vaidya.ai/vaidya/chat/completions \
  --header "authorization: Bearer $VAIDYA_API_KEY" \
  --header "content-type: application/json" \
  --data '{
  "model": "Vaidya-v2",
  "messages": [
    {
      "role": "user",
      "content": "Explain the pathogenesis of rheumatoid arthritis."
    }
  ]
  "temperature": 0.2
}'

Same body as JSON:

{
  "model": "Vaidya-v2",
  "messages": [
    {
      "role": "user",
      "content": "Explain the pathogenesis of rheumatoid arthritis."
    }
  ]
  "temperature": 0.2
}

Error handling scenarios

Failures return JSON with an error object (message, type, code) and an HTTP status, similar to OpenAI-style APIs. Typical situations:

Scenario	Status	What to do
No `Authorization` header, wrong key format, or invalid key	401	Send `Authorization: Bearer <key>` with a key from console.vaidya.ai. See Authentication.
Key revoked or expired	401	Create a new key in the console and update your config.
Key lacks access to the resource or project	403	Check account or project permissions in the console.
`messages` missing roles, empty content, or invalid structure	400	Match the chat schema: alternating user/assistant turns with string `content` (or valid multimodal parts).
File upload missing when required	400	Attach the required PDF or image.
Request or file payload too large	413	Shorten text, compress images, or split work across calls.
Field types or enums fail validation	422	Align types with the API reference (e.g. `temperature` as number).
Too many requests in a short window	429	Retry with exponential backoff; honor `Retry-After` when present.
Client or upstream timeout	408	Increase timeout, retry idempotent calls, or shorten prompts.
Transient server or overload	500 / 503	Retry with backoff; log the response/request id if you need support.

Who is this for?

Health apps - add symptom triage, drug info, or lab analysis to your product.
Wellness platforms - generate health scores and personalized plans.
Clinical tools - build copilots and workflow assistants for care teams.
Enterprise - embed healthcare AI into internal tools and customer-facing products.

Next steps

Get your API key at console.vaidya.ai.
Write better prompts - read the Prompt Guide.
Start building - jump to the Chat Completions API reference.
Ship safely - follow Best Practices for prompts, retries, security, and disclaimers.
Watch usage - use the Console Monitoring Usage page for quotas, charts, and alerts.

Introduction

On this page