Activity
Your organization's usage across models, apps, and users.
GMT+2
Filter
Total spend
$4,281
â–² 12.0%vs last month
Requests
304,630
â–² 18.4%vs last month
Token volume
214.8M
prompt + completion + reasoning
Cache hit rate
34.8%
â–² 4.1ppsaved $1,493
Blended $/M tokens
$1.42
â–¼ 6.8%vs last month
Usage
Requests over time · by model
GMT+2
Filter
Top users
By spend this month
- SCProductSarah Chensarah@acme.com$892
- MRProductMarcus Reidmarcus@acme.com$741
- DRPlatformDiego Ramirezdiego@acme.com$612
- HWProductHannah Weisshannah@acme.com$388
- LKProductLena Kowalskilena@acme.com$340
- PMOfficePriya Menonpriya@acme.com$187
Top apps
Where the traffic comes from
- Chat168,430 req$2,41255.3%
- Coding CLI74,228 req$1,10824.4%
- API (direct)48,891 req$61916.0%
- Knowledge (RAG)13,081 req$1424.3%
Usage type
Own shelf vs BYOK
Own shelf
74.2%
$3,177 · Llama, gpt-oss, embed
BYOK
25.8%
$1,104 · governed passthrough
OpenAI
$682
Anthropic
$422
Token breakdown
214.8M tokens this month
- Prompt143.2M66.7%
- Completion51.9M24.2%
- Reasoning19.7M9.2%
Semantic caching
34.8% hit rate · saved $1,493 this month
Cached: 74.8MUncached: 140.0M
By model
Llama 3.3 70B
41.2%$682
gpt-oss-120b
32.1%$420
claude-4-sonnetBYOK
28.7%$288
voyage-3-embed
15.3%$103