Infinity API
Hong Kong & U.S. customers only
Three-site AI Gateway

Dual-Entry AI Gateway

API Access · Online Chat · Unified Billing

Developers can use site-issued API keys with openclaw, Hermes, or other OpenAI-compatible clients; users can chat with AI directly in the browser.

Available only to customers in Hong Kong and the United States.

openclaw / Hermes compatible In-browser ChatGPT-style chat Multi-site smart routing
Animated three-site unified gateway diagram
Unified Gateway Smart Routing
Site A Mainstream Models
Site B Global Resources
Site C High Value
Dual entry

Two Ways to Use Infinity API

One account, one balance, and one key set for both client-side API access and web chat.

API

Client API Key Access

For openclaw, Hermes, Immersive Translate, Cherry Studio, and other OpenAI-compatible tools.

Base URL https://your-domain.com/v1
Authorization Bearer your Infinity API key
Chat Path /chat/completions
Chat

In-Browser AI Chat

Users can chat directly in the browser without configuring a client.

Multi-Model Routing

Automatically selects the best site and model for stable, fast, low-latency requests.

API Key Management

Manage multiple keys with permissions, groups, and usage limits.

Balance and Billing

Real-time balance, usage-based billing, transparent statements, and recharge approval.

Request Logs

Full request logs, monitoring, search, export, and alerts.

Plan Packaging

Flexible plan setup with multiple billing cycles and delivery workflows.

High-Availability Gateway

Multi-region failover with a 99.9% availability target.

Balance Recharge · Usage-Based Billing

Recharge First, Pay by Real Token Usage

Monthly, quarterly, and yearly plans are service-period benefits, not unlimited usage. Each successful request deducts balance based on real tokens and model pricing.

Monthly 30-day access

For light usage. Keeps account access, API key, web chat, and billing records; each call deducts balance by real token usage.

Quarterly 90-day access

For steady usage. Longer service period; balance works for both API access and web chat.

Yearly 365-day access

For long-term team usage with manual order approval, key reissue, and customer controls.

Recharge Options

Choose a Balance Package by Budget

Recharge amounts go into the account balance. Daytime usage uses the standard rate, while off-peak cards apply discounted limited tokens after 22:00.

Starter￥99For testing and light chat
Popular￥188For personal API access
Recommended￥388For steady daily usage
Advanced￥688For high-frequency chat and tool access
Team￥1000For shared team balance
Off-Peak Discount Card After 22:00 Limited discounted tokens for late-night usage, helping reduce peak-time gateway pressure.
Stable and Reliable Infrastructure
128ms

Avg. latency

99.95%

Availability

12+

Available regions

200+

Supported models

Live status: All systems operational
Model Market

Only Two GPT Chat Models Are Available

Client API access and web chat now use only gpt5.4 and gpt5.5.

Public Dashboard

Live Platform Status

The public page only shows traffic and reliability information. Personal tokens, charges, billing, and docs live in the logged-in workspace.

Live users 0 Real active visitors now
New users today 0 Real new accounts today
Daily token usage 0 Real token usage today
Order mode Manual approval Recharge and access are confirmed by admin

7-Day User Login Trend

Only real server login stats are shown; no activity stays at 0.

System status All gateways operational Live checks · multi-site failover
Coverage 12+ regions Major global regions
Model support 200+ models Mainstream, vision, reasoning, embeddings
Account Login

Login Infinity API

No account yet?
Account Sign Up

Sign up Infinity API

Already have an account?