DeepSeek, Qwen, GLM and Kimi alongside GPT, Claude and Gemini — all behind one OpenAI-compatible endpoint.
One balance and one bill for Chinese and global models. Change one line of base_url to switch over. Sign up with an email; no Chinese phone number needed.
Popular Chinese and global models for chat, code, image generation and embeddings.
Start calling every model in three steps.
Sign up with your email and create an API key in the dashboard. Minimum top-up $5, pay as you go.
Change the base URL to XHuoAPI's endpoint — OpenAI SDK users only need to modify one line.
Access 70+ models such as DeepSeek, Qwen, Claude and GPT through one unified API. Pay as you go.
curl --request POST \
--url https://api.xhuoapi.ai/v1/chat/completions \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data '{
"model": "gpt-4o",
"messages": [
{
"role": "system",
"content": "You are a professional AI assistant."
},
{
"role": "user",
"content": "Walk me through the history of artificial intelligence."
}
]
}'One platform, transparent pricing, real support — so you can focus on shipping.
Each model is served by multiple upstream pools. Four tiers trade price against stability, and traffic can be switched when a pool degrades.
Pay as you go, with per-token pricing published for every model. Four tiers range from highest stability to lowest cost. No monthly fee, no hidden charges.
Keys, usage stats & quota control — all models managed in one dashboard.
Fully OpenAI-compatible — switch your existing code with a single config change.
Cross-convert to OpenAI, Claude & Gemini compatible formats. Flexible framework integration.
Send integration or billing questions to support@xhuoapi.ai. We usually reply within one business day.
Supported Model Providers
Quick answers to key questions about XHuoAPI.
XHuoAPI is an AI model gateway. One OpenAI-compatible endpoint gives you 70+ models — Chinese models such as DeepSeek, Qwen, GLM, Kimi and Doubao alongside GPT, Claude, Gemini and Grok. One key, one bill, one usage dashboard.
70+ models across chat (DeepSeek, Qwen, GLM, Kimi, Doubao, GPT, Claude, Gemini, Grok), image generation, embeddings and text-to-speech — 9 providers in total, all through one unified API.
No. XHuoAPI accounts are created with an email address — no mainland China phone number or bank card. Some vendors also run international sites for overseas users, so if you only need one vendor, signing up there directly works too. XHuoAPI is for when you want one key across several vendors.
Simple: 1) Sign up and get an API key (no credit card required); 2) Change your base_url to XHuoAPI's endpoint; 3) Call using OpenAI-compatible format. For existing OpenAI SDK projects, it's just a one-line change.
Each API key can have its own spending limit, expiry date and allowed models, set in the console. If you suspect a key has leaked, delete it and create a new one at any time. All requests go over HTTPS.
If you only use one provider, going to that provider directly is usually simplest. XHuoAPI solves the cross-provider problem: one account, one bill and one usage dashboard across 9 providers, instead of registering and maintaining keys on each platform and dealing with each one’s sign-up and payment requirements separately.
One OpenAI-compatible endpoint · 9 model providers