Cost & Usage Observability
All Providers Operational
Total Spend (30d)
$3,842.18
Across 4 providers, 12 models
Token Volume
412.8M
308.2M prompt / 104.6M completion
P95 Latency & TTFT
740ms
195ms TTFT | 0.14% error rate
Cache Efficiency
$614.30
28.4% prompt cache hit rate
Model Usage & Cost Trends
Daily token consumption and cost breakdown by model
Provider Spend Allocation
Cost distribution and token rates across providers
OpenAI$2,074.78
54% · $0.032/1K tokens
Anthropic$1,191.08
31% · $0.028/1K tokens
Google Cloud$422.64
11% · $0.018/1K tokens
Groq$153.68
4% · $0.004/1K tokens
Input vs Output Ratio
75%/25%
308.2M prompt
104.6M completion
1,240Avg Prompt Length
413Avg Completion Length
Prompt cache: 28.4%
Ratio: 3.0:1Latency Timeline
610ms
P99
P90
P50
Request Status
1,534Total requests (24h)
200OK
1,288occurrences
429Rate Limit
219occurrences
500Server Error
27occurrences
API Traces
Monitor and analyze your recent API requests.
| Trace ID | Status | Time | Model | Tokens | Latency | Cost | Action |
|---|---|---|---|---|---|---|---|
| 200 | Just now | GPT-4oOpenAI | 977 / 417 | 816ms98ms TTFT | $0.00661 | ||
| 200 | 4m ago | Claude 3.5 SonnetAnthropic | 1,106 / 305 | 824ms144ms TTFT | $0.00698 | ||
| 200 | 8m ago | Gemini 1.5 ProGoogle Cloud | 428 / 577 | 637ms59ms TTFT | $0.00342 | ||
| 200 | 12m ago | Llama 3.3Groq | 1,666 / 313 | 632ms195ms TTFT | $0.00079 | ||
| 200 | 16m ago | GPT-4oOpenAI | 1,279 / 150 | 380ms171ms TTFT | $0.00470 | ||
| 200 | 20m ago | Claude 3.5 SonnetAnthropic | 1,996 / 405 | 603ms332ms TTFT | $0.01085 | ||
| 200 | 24m ago | Gemini 1.5 ProGoogle Cloud | 473 / 214 | 219ms332ms TTFT | $0.00166 | ||
| 429 | 28m ago | Llama 3.3Groq | 1,419 / 354 | 425ms154ms TTFT | $0.00077 | ||
| 429 | 32m ago | GPT-4oOpenAI | 215 / 285 | 784ms168ms TTFT | $0.00339 | ||
| 500 | 36m ago | Claude 3.5 SonnetAnthropic | 390 / 183 | 0ms0ms TTFT | $0.00337 |
Showing 1 to 10 of 15 traces