LLM API

Unified interface for calling large language models across multiple providers through a single endpoint.

What It Does

The LLM API abstracts provider differences so you can switch between OpenAI, Anthropic, Google, and Grok without changing your code. One endpoint, one format, any model.

Key Capabilities

Capability	Description
Multi-Provider	OpenAI, Anthropic, Google, Grok — all through one API
Auto-Routing	Set model: "auto" and let the system pick the best model for your task
Structured Output	Force JSON responses with schema validation
Multimodal	Send images alongside text — auto-converted to each provider's native format
Streaming	Real-time token-by-token responses via SSE
Global Cache	Automatic response caching for up to 40% cost reduction

Quick Example

curl -X POST https://llm.zihin.ai/api/v3/llm/public/call \
  -H "X-Api-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "Summarize this contract", "model": "auto"}'

Endpoints

Endpoint	Auth	Description
POST /api/v3/llm/public/call	API Key	LLM call (external integrations)
POST /api/v3/llm/call	JWT	LLM call (multi-tenant frontend)
GET /api/v3/llm/models	Public	List available models
POST /api/v3/llm/test-connection	JWT	Test provider connectivity

Next Steps

• Making Calls — Request/response format, structured output, multimodal
• Models & Tiers — Available models and plan tier access
• Streaming — Real-time SSE streaming
• Secrets & BYOK — Use your own provider API keys