Flat subscription · transparent per-window limits

Codex CLI. Done right.

OpenAI-compatible Codex gateway for indie developers. Transparent limits, fair price.

Sign up freeAPI docs

Subscription plans

Five tiers — billed monthly, transparent monthly message limits

Free Trial

$0 — $3 credit on signup · 3-day access · GPT-5.3-Codex

Standard

Standard

$10/mo
Monthly1,000 messages
Capacity2× Plus
  • Multi-channel routing
  • 30 req / min
  • All major models
  • Smart cache (up to 90% savings)
  • Good for solo developers
Get started
Most Popular
Pro

Pro

$25/mo
Monthly2,500 messages
Capacity5× Plus
  • High-availability routing
  • 80 req / min
  • 24 / 7 uptime guarantee
  • Priority request queue
  • Smart cache (up to 90% savings)
  • All major models
  • Good for small teams
Get started
Flagship

Flagship

$55/mo
Monthly6,000 messages
ValuePro $100 at half price
  • Top-priority routing
  • 200 req / min
  • 24 / 7 high availability
  • Highest priority queue
  • Smart cache (up to 90% savings)
  • Early access to new models
  • Dedicated support (2 hr response)
  • Good for heavy users and teams
Get started
NEW
Ultra

Ultra

$120/mo
Monthly13,500 messages
ValuePro $200 at 40% off
  • Max-priority routing
  • 500 req / min
  • 24 / 7 high availability
  • Dedicated support (2 hr response)
  • Early access to new models
  • Smart cache (up to 90% savings)
  • Best for power users & teams
Get started

All paid plans use weighted message count. gpt-5.4 medium / mini messages cost less than gpt-5.2 xhigh.

Message counts shown as gpt-5.2 xhigh equivalent.

Pay-as-you-go · Direct routing

Enterprise-readyComing soon
  • ✓ Native upstream API — no request parameter modification, no data storage
  • ✓ Unlimited calls, no monthly quota cap, elastic scaling
  • ✓ Value-add pricing: official rate + network optimization and dedicated support
  • ✓ 99.9% SLA
  • ✓ Full audit trail — meets enterprise compliance requirements
This feature is in development. Check back soon.

Need cost control? Use a subscription plan. Need unlimited calls? Pay-as-you-go is coming soon.

Model pricing

ModelInput / 1M tokensCached input / 1M tokensOutput / 1M tokens
GPT-5.4$2.50$0.25$15.00
GPT-5.4-mini$0.75$0.075$4.50
GPT-5.2$1.75$0.18$14.00
GPT-5 / 5.1$1.25$0.125$10.00

Billing examples

Based on subscription plan rates. Estimates for reference.

Quick Q&A

Scenario1,000 input + 300 output tokens
ModelGPT-5.4-mini
≈ $0.002
A few cents per thousand questions

Code fix

Scenario8,000 input + 2,000 output tokens
ModelGPT-5.4
≈ $0.05
Under a nickel per bug fix

Long-context reuse

Scenario20,000 cached + 2,000 input + 1,000 output
ModelGPT-5.4
≈ $0.025
Cache reuse saves ~80% vs no cache

FAQ

Start with ClaudeZH

Connect to Codex CLI in under 5 minutes.

Sign up free