KavTora API router docs

KavTora is an all-in-one AI SaaS platform and AI API router with a simple chatbot, one API key, and access to GPT, Claude, Gemini, DeepSeek, image, video, audio, embeddings, usage tracking, fallback routing, and one account. Use https://kavtora.com/api/v1 as the base URL for OpenAI-compatible requests. Create a demo key from the workspace, then configure upstream provider credentials in Vercel for live routing.

Quick request

POST https://kavtora.com/api/v1/chat/completions
Authorization: Bearer $API_ROUTER_KEY
Content-Type: application/json
model: auto
fallback: enabled

Available routes

GET/api/v1/models

OpenRouter-style all-in-one AI model catalog.

POST/api/v1/key

Issue a demo API key when production key enforcement is off.

POST/api/v1/chat/completions

OpenAI-compatible chat completions route.

POST/api/v1/responses

OpenAI-compatible Responses API demo route.

POST/api/v1/embeddings

Embedding-compatible demo route.

GET/api/v1/usage

Authenticated usage summary for API review.

Production environment

AI_ROUTER_API_KEY enforces the production API key.
AI_ROUTER_UPSTREAM_URL should point to an OpenAI-compatible upstream base URL.
AI_ROUTER_UPSTREAM_API_KEY authorizes live upstream provider traffic.
Public account-auth environment variables keep login available on this domain.

1. Ask the AI assistant

Use this step to keep API routing explicit, reviewable, and aligned with provider policy, billing records, and production fallback needs.

2. Choose a model

Use this step to keep API routing explicit, reviewable, and aligned with provider policy, billing records, and production fallback needs.

3. Connect the API

Use this step to keep API routing explicit, reviewable, and aligned with provider policy, billing records, and production fallback needs.