Documentation
Everything you need to build with the LLMAI API gateway.
Getting Started with LLMAI
LLMAI is an AI model gateway that gives your application access to dozens of models across leading model families — including GLM 5.3 Flash; GPT-4.1, GPT-4o, GPT-5, GPT-5 Mini and the GPT-5.6 family; Claude Opus, Sonnet, and Haiku; Gemini 2.5 and 3.x; DeepSeek R1 and V4; plus Kimi, MiniMax, Qwen, Gemma, MiMo, Grok, and Veo. All are available through a single OpenAI-compatible HTTP endpoint.
Save up to 70% on published API rates. GLM 5.3 Flash is $0.045 input, $0.009 cached input, and $0.150 output per million tokens.
🎁 Free $2 trial credit available on request. Not ready to deposit? Contact us to request trial credit and test any model before you commit.
What's in These Docs
| Section | What You'll Learn |
|---|---|
| Quickstart | Register, fund your account, and fire your first request |
| API Overview | Endpoint URLs, authentication headers, request/response schemas |
| Models | Full catalog of supported models with slugs, capabilities, and pricing |
| Billing | How prepaid balances work and how deductions are calculated |
| Integrations | Step-by-step setup for SDKs, CLI tools, and editors |
| Troubleshooting | Diagnose and resolve the most common API errors |
Your First API Call
curl https://api.llmai.dev/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "kimi-k2.7-code",
"messages": [{"role": "user", "content": "Hello!"}]
}'At a Glance
- Endpoint:
https://api.llmai.dev/v1 - Authentication:
Authorization: Bearer <your-key> - Protocol: OpenAI Chat Completions format — fully compatible with existing SDKs
- Payment model: Prepaid balance, deducted per request — no monthly subscriptions
- Cost: Save over 65% vs going direct to providers
Head to the Quickstart to be up and running in minutes.
Community
Have questions, want to share what you're building, or need a hand? Join the LLMAI community — the team and other developers are there.