Pay only for what you use

Start free with real models. No credit card required. Scale seamlessly with transparent per-token pricing.

$5 free creditsNo card requiredMicro-cent billing

Evaluation

Explore the platform with real models

$0/forever
  • All standard models
  • Intelligent routing (routeplex-ai)
  • Daily usage limits apply
  • OpenAI SDK compatible
  • Real-time cost tracking
  • Same API, same endpoints
  • Premium models
  • Priority support

No credit card required. No sandbox — evaluation limits apply automatically.

Start Free
Most Popular

Pay As You Go

Full access, billed per token

Usage/based
  • Micro-cent billing per token
  • All standard + premium models
  • Intelligent routing (routeplex-ai)
  • Budget controls & cost alerts
  • 10,000 RPM rate limit
  • Real-time usage dashboard
  • Automatic failover
  • Email support

Usage billed at account level. Real-time cost display. Non-refundable.

Get Started

Enterprise

Volume pricing, SLA, dedicated support

Custom/pricing
  • Volume discounts
  • All models included
  • Custom rate limits
  • SLA guarantee
  • Dedicated account manager
  • Custom integrations
  • High-risk use case approval
  • Priority incident response

Medical, legal, financial, and government use cases require enterprise approval.

Contact Sales

Model pricing

Per 1M tokens. Use intelligent routing to automatically select the most cost-effective model, or specify directly.

OpenAI

GPT-4o
In: $2.50Out: $10.00
GPT-4o Mini
In: $0.15Out: $0.60
GPT-4 TurboPremium
In: $10.00Out: $30.00

Anthropic

Claude Sonnet 4
In: $3.00Out: $15.00
Claude 3.5 Haiku
In: $0.80Out: $4.00
Claude Opus 4Premium
In: $15.00Out: $75.00

Google

Gemini 2.0 Flash
In: $0.075Out: $0.30
Gemini 1.5 Pro
In: $1.25Out: $5.00

Prices per 1M tokens. Rates may differ from upstream provider pricing. Models may change without notice in auto routing mode. View all models

Real-time cost tracking

Micro-cent precision. Daily limits and budget alerts built in.

99.9% effective uptime

Multi-provider failover. Requests silently reroute.

Stateless by design

Zero prompts stored. In-memory processing only.

Encrypted end-to-end

TLS 1.3 in transit. API keys hashed. GDPR-aligned.

Frequently asked questions

Ready to start building?

Free evaluation plan with real models. Upgrade anytime.