⚡ Streaming · Multi-subscription · Auto-discovery

One gateway.
Every model. Zero friction.

High-speed access to China's top AI models with streaming SSE, multi-subscription failover, automatic model discovery and 1M-token context.

Start free — 100K tokens/mo Read the docs
curl 
curl https://api.silkgateway.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": true
  }'
Models
—
auto-discovered
Context
1M
input tokens
Max output
128K
tokens by default
Edge
300+
Cloudflare locations

Built for production traffic

Everything you need between your app and the models.

🌊
Streaming SSE
Real-time token streams for OpenAI and Claude formats, with correct event framing the official SDKs expect.
🔀
Multi-subscription failover
Aggregate many subscriptions behind one model name. On 429 the gateway silently switches to the next source — your users never notice.
🔍
Auto model discovery
One click pulls the upstream model list into your gateway config. New models are routable instantly.
📐
1M in · 128K out
Long-context by default. If a model caps lower, the gateway auto-adjusts and retries transparently.
🔁
OpenAI ↔ Claude
Claude SDK and Claude Code work out of the box via /v1/messages — full bidirectional format translation, streaming included.
🗜️
RTK Token Saver
Tool outputs (git diff, grep, builds) are compressed before forwarding — typically 20-40% input token savings.

Models & pricing

Live from the gateway. Add your own providers and subscriptions anytime.

Model Input Output Context Provider

Prices per 1M tokens in USD. Combo names (e.g. fast-cheap) can also be used as the model field.

Up and running in minutes

Point your existing OpenAI/Claude client at the gateway. That's it.

1
Create an account
Sign up with email, verify, and get your API key instantly.
2
Switch your base URL
Set base_url to https://api.silkgateway.ai/v1 and use your key. Works with OpenAI SDK, Claude SDK, LangChain, Claude Code.
3
Ship
Streaming, failover and billing just work. Watch usage live in the dashboard.

Simple pricing

Start free. Upgrade when your traffic grows.

Free
$0
  • ✓100K tokens / month
  • ✓10 requests / min
  • ✓All models & streaming
Start free
Pro
$29/mo
  • ✓1M tokens / month, then pay-as-you-go
  • ✓200 requests / min
  • ✓10% usage discount
Choose Pro
Enterprise
Custom
  • ✓10M tokens / month
  • ✓2000 requests / min
  • ✓20% usage discount
Contact us