Pricing
Pay for routing. Save on inference.
Thalera charges a flat fee per routed call. You keep your own provider accounts and your own rates — we just make sure the cheapest capable model answers.
Free
$0
Up to 50k routed calls / month
- Cost-optimized routing
- Semantic cache (1k entries)
- 7-day request logs
- Community support
Team
Popular$0.20
per 1k routed calls, after the free tier
- Custom routing policies per key
- Unlimited cache with tenant namespaces
- 90-day logs and cost analytics
- Budget alerts and spend caps
- Cross-provider failover
- Email support, 1 business day
Enterprise
Custom
Volume pricing and committed capacity
- Dedicated routing region or VPC
- SSO, audit logs, SOC 2 report
- Self-hosted classifier option
- 99.99% SLA with support escalation
- Solution engineer on onboarding
Savings calculator
A rough estimate against sending every call to a frontier model. Uses published list prices and a typical 17% cache hit rate.
1M
1.2K
15%
Estimated monthly
- Frontier-only baseline
- $10,800
- Inference with Thalera routing
- $2,152
- Thalera routing fee
- $200
- You save
- $8,448
78% lower than frontier-only
Illustrative only. Actual savings depend on your traffic mix and provider rates.