Token & Cost
Usage
Agent Evals
All sources
OpenClaw
Hermes
All contexts
Group chat
Direct chat
Cron
CLI / TUI
Subagent
Refresh
Run evals
Filters ▾
Today
Yesterday
Last 2 days
7 days
30 days
90 days
All time
Custom:
→
Apply
Clear
Cost by day
Cost by model
Cost by agent
Cost by chat context
OpenRouter billing
Loading…
Cost anomalies
2× 7-day baseline
Agent cost ranking
Unpriced sessions
Model ROI
cost · turns · sessions
Idle sessions
>20k tokens, ≤2 turns
Cache efficiency
Evals cost by context
Sessions
Source
Session
Model
Tokens
Cost $
Updated
Agent Evals leaderboard
Model
Pass rate
Per-type breakdown
Tasks
Runs
Avg latency
Avg tokens
Avg cost
Total cost
Run
Loading…
Manual runs only — nothing fires on a timer. A single-task run is a few cents; a full sweep runs every task × the chosen model.
Tasks
Run now · full sweep
+ New task
Loading…
Run history
All results
PASS
FAIL
Loading…