LIVE · Singapore · <5ms latency · 99.9% uptime

One API for
All AI Models

Call GPT-5, Claude 4, Gemini, DeepSeek — with one URL change.
No more 17 different API keys. Singapore edge, Redis cached, Asia-friendly payments.

No VPN · OpenAI-compatible · Free to start · WeChat Pay & Alipay accepted

Get free API key → Read the docs
Python · OpenAI SDK
# Before — slow from Asia, expensive
base_url = "https://api.openai.com/v1"
# After — Singapore edge, cached, fast
base_url = "https://aperture-ai.run/proxy/openai/v1"
✓ Same SDK ✓ Redis cached ✓ Auth managed ✓ Rate limited
<5ms
Proxy latency in Asia
80%
Cost reduction via cache
17+
AI providers unified
99.9%
Uptime SLA (Pro)
Multi-Provider

All major models. One endpoint.

Switch providers without changing your integration.

AI
OpenAI
/proxy/openai
热门
GPT-5 GPT-4.5 GPT-4o GPT-4o mini o3 o4-mini DALL·E 3 Whisper Embeddings v3
An
Anthropic
/proxy/anthropic
Claude 4 Opus Claude 4 Sonnet Claude 3.7 Sonnet Claude 3.5 Haiku Claude 3 Opus
G
Google Gemini
/proxy/gemini
Gemini 2.5 Pro Gemini 2.5 Flash Gemini 2.0 Flash Gemini 1.5 Pro
DS
DeepSeek
/proxy/deepseek
DeepSeek-V3 DeepSeek-R1 DeepSeek-R1-0528 DeepSeek-Coder-V2 DeepSeek-VL2
𝕏
xAI Grok
/proxy/xai
Grok-3 Grok-3 mini Grok-2 Vision Grok-2-1212
阿
Alibaba Cloud
/proxy/aliyun
Qwen3-235B Qwen-Max Qwen-Plus Qwen-VL-Max Qwen-Audio
Gq
Groq
/proxy/groq
Llama 4 Scout Llama 3.3 70B Mixtral 8x7B
Ki
Kimi
/proxy/moonshot
kimi-k2 moonshot-v1-128k moonshot-v1-32k
智
Zhipu AI
/proxy/zhipu
GLM-4-Plus GLM-4-Air GLM-4V-Flash
百
Baidu ERNIE
/proxy/baidu
ERNIE-4.5 ERNIE-4.0 ERNIE Speed
Mi
Mistral
/proxy/mistral
Mistral Large Codestral
Tg
Together AI
/proxy/together
Llama 4 Maverick Llama 4 Scout
Px
Perplexity
/proxy/perplexity
Sonar Pro Sonar Huge
Mx
MiniMax
/proxy/minimax
MiniMax-Text-01 abab6.5s
+2
StepFun · 01.ai
stepfun · lingyiwanwu
Step-2 Yi-Lightning
Quick Start

Up and running in 2 minutes

No SDK changes. No new libraries. Just update one line.

1

Get your API key

Register free, choose a plan, and copy your Aperture API key from the dashboard.

2

Change the base URL

Point your SDK to aperture-ai.run/proxy/{provider}/v1 — that's the only change needed.

3

Enjoy Singapore speed

Requests route through our Singapore edge. Cached responses return instantly — zero upstream cost.

Python Node.js cURL
from openai import OpenAI
client = OpenAI(
api_key="sg-xxxxxxxxxxxxxxxx",
base_url="https://aperture-ai.run/proxy/openai/v1"
)
response = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": "Hello!"}]
)
Features

Enterprise-grade infrastructure

Everything you need to ship AI products in Asia-Pacific.

⚡

Redis Response Cache

Identical prompts return from cache in microseconds. Slash API costs by up to 80% automatically with configurable TTL.

🌏

Singapore Edge Node

Co-located in Singapore for the lowest round-trip latency across SEA, East Asia, and Oceania. <5ms proxy overhead.

🔑

API Key Auth

Issue proxy keys to your team or customers. Upstream provider keys never leave the server — zero exposure risk.

🛡️

Rate Limiting

Distributed rate limiting via Redis. Prevent runaway jobs from blowing your upstream quota. Per-key limits on Pro.

📊

Usage Dashboard

Real-time visibility into requests, cache hit rate, latency, and per-provider usage — all in one place.

💳

Asia-Friendly Payments

Pay with Visa, WeChat Pay, or Alipay via Stripe. No overseas card required. Subscribe monthly or top up credits.

Pricing

Simple, transparent pricing

Pay with Visa · WeChat Pay · Alipay — via Stripe

VISA WeChat Alipay
Free
$0
/ month · forever
  • ✓100 requests / day
  • ✓OpenAI + Anthropic access
  • ✓60s response cache
  • ✗No SLA
  • ✗Community support only
Get started free
Most Popular
Pro
$19
/ month · billed monthly
  • ✓10,000 requests / day
  • ✓All providers incl. Alibaba
  • ✓5-min response cache
  • ✓99.9% uptime SLA
  • ✓Email support
Subscribe — $19/mo
Enterprise
$99
/ month · billed monthly
  • ✓Unlimited requests
  • ✓All providers
  • ✓Custom cache TTL
  • ✓99.99% uptime SLA
  • ✓Priority 24/7 support
Subscribe — $99/mo

All plans include Alipay & WeChat Pay · Cancel anytime

Infrastructure

Built for Asia-Pacific

Singapore-hosted for the lowest latency to the largest AI market.

🇸🇬
Singapore (Primary)
SEA, East Asia, ANZ · <5ms proxy overhead
Live
🇯🇵
Japan (Coming Soon)
Japan, Korea
Soon
🇦🇺
Australia (Coming Soon)
ANZ region
Soon
Latency comparison — Singapore developer
Direct to OpenAI US ~280ms
Via Aperture (live call) ~285ms
+5ms proxy overhead, full caching & auth
Cache hit (repeat prompt) <8ms 🚀
35× faster · zero upstream cost
Uptime (30-day rolling) 99.97%
TLS 1.3 · Zero-log policy · GDPR compliant

Start building faster
in Asia-Pacific

Free tier · No credit card · 2 minutes to integrate

Get free API key → Read the docs