Cut your Claude Code
costs by 95%
Drop-in proxy that automatically routes simple requests to cost-efficient models and complex ones to premium. One line to set up.
export ANTHROPIC_BASE_URL=https://api.inferroute.ai/v1Features
Everything you need, nothing you don't
Intelligent Routing
Automatically classifies request complexity in real-time. Simple tasks go to Haiku, complex reasoning goes to Opus. No configuration needed.
Drop-in Compatible
Set one environment variable and you're done. Works with Claude Code, any Anthropic SDK, and every tool that supports ANTHROPIC_BASE_URL.
Real-time Dashboard
Track every request, see exactly how much you're saving, and monitor model routing decisions — all in a live dashboard.
How it works
Three steps. That's it.
Set the URL
Point your Anthropic client at Inferroute. One environment variable, nothing else changes.
export ANTHROPIC_BASE_URL=https://api.inferroute.ai/v1Code normally
Use Claude Code exactly as before. Inferroute intercepts and classifies each request in real-time.
Save ~95%
Simple tasks get routed to cheaper models. Complex reasoning stays on premium. You save automatically.
Pricing
Pay per token. No surprises.
Plans include a monthly credit budget to spend at the rates below.
Free
- 2 API keys
- Cost-efficient
- Community support
Pro
- 10 API keys
- All models
- Priority support
Max
- Unlimited API keys
- All + priority routing
- Dedicated support
Need more? Top up your balance anytime from $5. Overages are billed at the rates above. Pay with card or USDC on Base (near-zero fees).
Ready to cut your costs?
Join developers who are saving thousands on their Claude Code bills. Set up in under a minute.
Free tier included. No credit card required.