Analytics
Spend, latency, reliability, and routing efficiency across your AI traffic.
Spend over time
Requests over time
Latency p50 / p95 by provider
Cost by model
Error rate
0.42%
Fallback rate
3.1%
Cache hit
18.4%
Cost / 1K req
$3.88
Top users / teams
| Model | Requests | Spend |
|---|---|---|
| openai/gpt-6-sol | 184,320 | $1240.55 |
| anthropic/claude-sonnet-5 | 142,880 | $980.10 |
| google/gemini-3.1-pro | 78,110 | $540.32 |
| groq/llama-4-scout | 52,640 | $84.20 |
| deepseek/v4-pro | 31,040 | $160.88 |