v1.0 — Now in Beta

One API. Every AI Model.

OpenAI-compatible endpoint. Credit-based billing. Smart routing across providers.

16+
models available
<200ms
average latency
99.9%
uptime

Start in 30 seconds

bash
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
# Install the SDK
"docs-token-command">npm install @conflux/router

# Or use any OpenAI-compatible client
"docs-token-command">curl https://theconflux.com/v1/chat/completions \
  -H "Authorization: Bearer cf_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "task_type": "code_gen",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

# Or specify a model directly
"docs-token-command">curl https://theconflux.com/v1/chat/completions \
  -H "Authorization: Bearer cf_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

It's just the OpenAI API. Swap the base URL and key.

Why developers switch

🔌

Drop-in Replacement

Works with the OpenAI SDK, LangChain, Vercel AI SDK, and any OpenAI-compatible client. Change 2 lines of code.

💰

Transparent Pricing

Pay only for what you use. Credits never expire. No monthly minimums. See exact costs before you call.

Smart Routing

Automatic failover between providers. Route by cost, speed, or quality. Your app never goes down.

Ready to build?

Free tier includes 500 credits. No credit card required.