GPT-5.5new
openai/gpt-5.5
−80%
5× cheaper
- Input / Mtok
- $5.00
- $1.00
- Output / Mtok
- $30.00
- $6.00
1.1M contextcache read · $0.100
Founding-developer pricing — live now
OpenAI tokens at outlet prices.
GPT-5.5, GPT-5, and GPT-5 Codex — billed at 20% of OpenAI list. Same SDK. One line change. Pay-as-you-go credits.
OpenAI list price
GPT-5.5
Input
$5.00/M
Output
$30.00/M
1.1M context
vs↓
On corouter
GPT-5.5
Input
$1.00/M
Output
$6.00/M
1.1M context
−80%
5× cheaper · same model
How it works
One endpoint. Every GPT, billed in outlet pricing.
Point the OpenAI SDK at api.corouter.co/v1 and route any GPT model by name. Claude and Gemini land next.
Your app
OpenAI SDK
corouter
20% of list
Live now · OpenAI
GPT-5.5
GPT-5
GPT-5 Codex
Coming soon
Anthropic
Gemini
Same SDK. Drop-in replace OPENAI_BASE_URL.
Pre-paid credits. $5 minimum. Live balance, transparent logs.
Cache passes through. OpenAI 90%-off cached input honored 1:1.
0%
Of OpenAI list price
0×
Cheaper than retail
$0
Dollar minimum to start
0%
Last-30-day uptime
Why corouter
Three reasons to swap your base URL.
Features
Built for the bill, not the brochure.
OpenAI-API-compatible, drop-in.
Point the OpenAI SDK at api.corouter.co/v1 and you're done. Streaming, tool calls, vision, JSON mode — every surface passes through unchanged.
from openai import OpenAI
client = OpenAI(
base_url="https://api.corouter.co/v1", # one line
api_key="cr-...",
)
# the rest of your codebase doesn't change
resp = client.chat.completions.create(
model="openai/gpt-5.5",
messages=[{"role": "user", "content": "ship it"}],
)Your bill, but smaller.
GPT-5.5 at 20% of list. Same model ID, same JSON, 5× cheaper output.
OpenAI list
$5.00 in · $30.00 out
corouter
$1.00 in · $6.00 out
−80% · 5× cheaper
GPT today. Claude and Gemini next.
We ship one vendor at a time, properly. Anthropic and Google land this summer at the same outlet pricing posture.
Claude family
Summer 2026Gemini family
Summer 2026Three GPT flagships live now.
GPT-5.5, GPT-5, and GPT-5 Codex. Each row shows OpenAI list with a line through it, and what you actually pay.
GPT-5.5
1.1M ctx
$30.00
$6.00
GPT-5
400K ctx
$10.00
$2.00
GPT-5 Codex
400K ctx
$10.00
$2.00
GPT-5.5
1.1M ctx
$30.00
$6.00
GPT-5
400K ctx
$10.00
$2.00
GPT-5 Codex
400K ctx
$10.00
$2.00
GPT-5.5
1.1M ctx
$30.00
$6.00
GPT-5
400K ctx
$10.00
$2.00
GPT-5 Codex
400K ctx
$10.00
$2.00
GPT-5.5
1.1M ctx
$30.00
$6.00
GPT-5
400K ctx
$10.00
$2.00
GPT-5 Codex
400K ctx
$10.00
$2.00
Bring a teammate. Both bills halve again.
Founding-developer referral: share a key, tap another 50% off your next $20 of usage. Built for word-of-mouth.
Your link
corouter.co/r/your-handle
Earnings this month$0 → unlocked at first signup
Zero data retention by default.
No prompt logging. Per-key audit log streamed to your S3 on request. Instant credit refund if a request fails to charge.
- Same OpenAI capacity — billed in plain dollars
- Live balance + per-request margin shown
- EU + US edge POPs, China-direct routing soon
Catalog
3 GPT family flagships live. Claude and Gemini next.
GPT-5.5
$30.00 $6.00 out / Mtok
GPT-5
$10.00 $2.00 out / Mtok
GPT-5 Codex
$10.00 $2.00 out / Mtok
GPT-5.5
$30.00 $6.00 out / Mtok
GPT-5
$10.00 $2.00 out / Mtok
GPT-5 Codex
$10.00 $2.00 out / Mtok
GPT-5.5
$30.00 $6.00 out / Mtok
GPT-5
$10.00 $2.00 out / Mtok
GPT-5 Codex
$10.00 $2.00 out / Mtok
GPT-5.5
$30.00 $6.00 out / Mtok
GPT-5
$10.00 $2.00 out / Mtok
GPT-5 Codex
$10.00 $2.00 out / Mtok
Claude family
Landing summer 2026
Gemini family
Landing summer 2026
Claude family
Landing summer 2026
Gemini family
Landing summer 2026
Claude family
Landing summer 2026
Gemini family
Landing summer 2026
Claude family
Landing summer 2026
Gemini family
Landing summer 2026
Quickstart
Three lines from any SDK.
Same OpenAI-compatible payload. Pick a model by slug — region routing happens at the gateway.
api.corouter.co/v1
TypeScriptPythoncurl
import OpenAI from "openai"
const client = new OpenAI({
baseURL: "https://api.corouter.co/v1",
apiKey: process.env.COROUTER_KEY,
})
// Same SDK, just route by model name.
const overseas = await client.chat.completions.create({
model: "anthropic/claude-opus-4.7",
messages: [{ role: "user", content: "Plan the launch" }],
})
const china = await client.chat.completions.create({
model: "deepseek/v4-pro",
messages: [{ role: "user", content: "Plan the launch" }],
})
FAQ
The questions every dev asks first.
$5 free
Stop paying retail for GPT.
Sign up, generate a key, drop in the OpenAI SDK. First $5 of credits on us — roughly 800K GPT-5.5 output tokens or 2.5M GPT-5 output tokens to play with.