Raw data
Logs
Every check our lab has saved: the good ones, the failed ones, and the tokens they cost us. This is the ledger behind every board on the site.
What our lab spent
Tokens the paid apps reported back to us for our own test runs. A token is a small chunk of text, about four letters. Total is new input + cache + output.
Pick a time window
Every count on this page follows the window you pick here. The default is the last 24 hours. 30d is as far back as we serve.
Total tokens · last 5 hours
462,417
62 samples · Claude API-eq $0.2462 · not our bill (see note)
How to read this
Cost is a price some apps report with each answer — what the same work would cost at API rates. It is not our bill; we pay flat monthly plans, and Codex reports no figure at all. Quotas are a different thing, and we cannot see them.
How to read this
New in is uncached prompt tokens. Cache is cache reads plus writes — cheaper, but still usage. Out is everything the model wrote, including think. Total adds those three. Bigger is not better here. It is what our own testing used.
By app
| Split | n | New in | Cache | Out | Total |
|---|---|---|---|---|---|
| Codex | 25 | 262.7k | 39.2k | 5,037 | 306.9k |
| Grok Build | 12 | 135.7k | 3,328 | 2,414 | 141.5k |
| Claude Code | 25 | 7,285 | 0 | 6,768 | 14.1k |
By check
| Split | n | New in | Cache | Out | Total |
|---|---|---|---|---|---|
| speed | 48 | 305.1k | 41.3k | 13.7k | 360.1k |
| ping | 14 | 100.6k | 1,152 | 484 | 102.3k |
By model
| Split | n | New in | Cache | Out | Total |
|---|---|---|---|---|---|
| GPT-5.6 Luna | 10 | 120.2k | 0 | 1,296 | 121.5k |
| Grok 4.6 | 8 | 87.3k | 2,176 | 1,288 | 90.7k |
| GPT-5.6 Terra | 5 | 38.2k | 23.6k | 1,226 | 63.0k |
| GPT-5.6 Sol | 5 | 46.1k | 15.6k | 1,215 | 63.0k |
| GPT-5.5 | 5 | 58.1k | 0 | 1,300 | 59.4k |
| Grok 4.5 | 4 | 48.5k | 1,152 | 1,126 | 50.8k |
| Haiku 4.5 | 10 | 2,610 | 0 | 3,109 | 5,719 |
| Sonnet 5 | 5 | 1,610 | 0 | 1,249 | 2,859 |
| Fable 5 | 5 | 1,535 | 0 | 1,205 | 2,740 |
| Opus 5 | 5 | 1,530 | 0 | 1,205 | 2,735 |
Test runs
6 launches in this window · 10 failed samples. Open a run to see every model it touched and what that cost.
Aug 21, 12:00:00 AMpulsescheduled1m 44s · 13 rec · 0 skip · 5 fail · 98.2k tok
| Target | Result | Wall | New in | Cache | Out | Total |
|---|---|---|---|---|---|---|
| pingclaude | rate limit warningserved with a rate limit warning: seven_day at 0.28 utilization | 6.91s | 250 | 0 | 130 | 380 |
| pingcodex | ok | 8.59s | 12.1k | 0 | 5 | 12.2k |
| pinggrok | ok | 6.14s | 9,691 | 256 | 18 | 9,965 |
| speedclaude/haiku | rate limit warningserved with a rate limit warning: seven_day at 0.28 utilization | 6.09s | 272 | 0 | 558 | 830 |
| speedclaude/sonnet | rate limit warningserved with a rate limit warning: seven_day at 0.28 utilization | 6.91s | 322 | 0 | 241 | 563 |
| speedclaude/opus | rate limit warningserved with a rate limit warning: seven_day at 0.28 utilization | 6.79s | 306 | 0 | 241 | 547 |
| speedclaude/fable | rate limit warningserved with a rate limit warning: seven_day at 0.28 utilization | 7.16s | 307 | 0 | 241 | 548 |
| speedcodex/luna | ok | 10.8s | 10.8k | 0 | 254 | 11.0k |
| speedcodex/terra | ok | 9.05s | 12.3k | 0 | 243 | 12.6k |
| speedcodex/sol | ok | 9.49s | 12.3k | 0 | 243 | 12.6k |
| speedcodex/gpt-5.5 | ok | 9.63s | 11.4k | 0 | 257 | 11.6k |
| speedgrok/grok-4.6 | ok | 6.92s | 12.1k | 256 | 294 | 12.7k |
| speedgrok/grok-4.5 | ok | 6.09s | 12.1k | 256 | 280 | 12.7k |
Aug 20, 11:00:02 PMpulsescheduled1m 35s · 10 rec · 3 skip · 5 fail · 64.2k tok
| Target | Result | Wall | New in | Cache | Out | Total |
|---|---|---|---|---|---|---|
| pingclaude | rate limit warningserved with a rate limit warning: seven_day at 0.26 utilization | 4.20s | 250 | 0 | 85 | 335 |
| pingcodex | ok | 9.19s | 12.1k | 0 | 5 | 12.2k |
| pinggrok | skipped · subscription auth missing | — | — | — | — | — |
| speedclaude/haiku | rate limit warningserved with a rate limit warning: seven_day at 0.26 utilization | 11.7s | 272 | 0 | 531 | 803 |
| speedclaude/sonnet | rate limit warningserved with a rate limit warning: seven_day at 0.26 utilization | 6.57s | 322 | 0 | 241 | 563 |
| speedclaude/opus | rate limit warningserved with a rate limit warning: seven_day at 0.26 utilization | 8.12s | 306 | 0 | 241 | 547 |
| speedclaude/fable | rate limit warningserved with a rate limit warning: seven_day at 0.26 utilization | 7.68s | 307 | 0 | 241 | 548 |
| speedcodex/luna | ok | 14.4s | 12.2k | 0 | 254 | 12.4k |
| speedcodex/terra | ok | 9.83s | 5,437 | 6,912 | 243 | 12.6k |
| speedcodex/sol | ok | 9.71s | 10.6k | 1,792 | 243 | 12.6k |
| speedcodex/gpt-5.5 | ok | 10.4s | 11.4k | 0 | 254 | 11.6k |
| speedgrok/grok-4.6 | skipped · subscription auth missing | — | — | — | — | — |
| speedgrok/grok-4.5 | skipped · subscription auth missing | — | — | — | — | — |
Aug 20, 10:00:00 PMpulsescheduled2m 8s · 13 rec · 0 skip · 0 fail · 101.0k tok
| Target | Result | Wall | New in | Cache | Out | Total |
|---|---|---|---|---|---|---|
| pingclaude | ok | 4.56s | 250 | 0 | 44 | 294 |
| pingcodex | ok | 8.67s | 12.2k | 0 | 5 | 12.2k |
| pinggrok | ok | 5.24s | 9,948 | 0 | 30 | 9,978 |
| speedclaude/haiku | ok | 10.3s | 272 | 0 | 552 | 824 |
| speedclaude/sonnet | ok | 7.47s | 322 | 0 | 241 | 563 |
| speedclaude/opus | ok | 7.45s | 306 | 0 | 241 | 547 |
| speedclaude/fable | ok | 8.41s | 307 | 0 | 241 | 548 |
| speedcodex/luna | ok | 14.9s | 12.2k | 0 | 254 | 12.4k |
| speedcodex/terra | ok | 12.4s | 9,535 | 2,816 | 243 | 12.6k |
| speedcodex/sol | ok | 15.8s | 5,439 | 6,912 | 243 | 12.6k |
| speedcodex/gpt-5.5 | ok | 12.9s | 12.8k | 0 | 267 | 13.0k |
| speedgrok/grok-4.6 | ok | 7.59s | 12.2k | 256 | 313 | 12.7k |
| speedgrok/grok-4.5 | ok | 8.33s | 11.9k | 512 | 283 | 12.7k |
Aug 20, 9:44:13 PMpulsecatch-up4.02s · 0 rec · 13 skip · 0 fail · 0 tok
| Target | Result | Wall | New in | Cache | Out | Total |
|---|---|---|---|---|---|---|
| pingclaude | skipped · already recorded this slot | — | — | — | — | — |
| pingcodex | skipped · already recorded this slot | — | — | — | — | — |
| pinggrok | skipped · already recorded this slot | — | — | — | — | — |
| speedclaude/haiku | skipped · already recorded this slot | — | — | — | — | — |
| speedclaude/sonnet | skipped · already recorded this slot | — | — | — | — | — |
| speedclaude/opus | skipped · already recorded this slot | — | — | — | — | — |
| speedclaude/fable | skipped · already recorded this slot | — | — | — | — | — |
| speedcodex/luna | skipped · already recorded this slot | — | — | — | — | — |
| speedcodex/terra | skipped · already recorded this slot | — | — | — | — | — |
| speedcodex/sol | skipped · already recorded this slot | — | — | — | — | — |
| speedcodex/gpt-5.5 | skipped · already recorded this slot | — | — | — | — | — |
| speedgrok/grok-4.6 | skipped · already recorded this slot | — | — | — | — | — |
| speedgrok/grok-4.5 | skipped · already recorded this slot | — | — | — | — | — |
Aug 20, 9:00:00 PMpulsescheduled1m 54s · 13 rec · 0 skip · 0 fail · 99.4k tok
| Target | Result | Wall | New in | Cache | Out | Total |
|---|---|---|---|---|---|---|
| pingclaude | ok | 4.62s | 250 | 0 | 62 | 312 |
| pingcodex | ok | 8.51s | 12.1k | 0 | 5 | 12.2k |
| pinggrok | ok | 6.50s | 9,436 | 512 | 30 | 9,978 |
| speedclaude/haiku | ok | 8.82s | 272 | 0 | 545 | 817 |
| speedclaude/sonnet | ok | 7.17s | 322 | 0 | 261 | 583 |
| speedclaude/opus | ok | 8.38s | 306 | 0 | 241 | 547 |
| speedclaude/fable | ok | 7.90s | 307 | 0 | 241 | 548 |
| speedcodex/luna | ok | 13.7s | 12.2k | 0 | 254 | 12.4k |
| speedcodex/terra | ok | 11.6s | 5,437 | 6,912 | 254 | 12.6k |
| speedcodex/sol | ok | 10.9s | 5,437 | 6,912 | 243 | 12.6k |
| speedcodex/gpt-5.5 | ok | 8.86s | 11.2k | 0 | 265 | 11.5k |
| speedgrok/grok-4.6 | ok | 7.20s | 12.2k | 256 | 293 | 12.7k |
| speedgrok/grok-4.5 | ok | 6.68s | 12.2k | 256 | 283 | 12.7k |
Aug 20, 8:00:00 PMpulsescheduled1m 54s · 13 rec · 0 skip · 0 fail · 99.6k tok
| Target | Result | Wall | New in | Cache | Out | Total |
|---|---|---|---|---|---|---|
| pingclaude | ok | 4.79s | 250 | 0 | 43 | 293 |
| pingcodex | ok | 8.96s | 12.1k | 0 | 5 | 12.2k |
| pinggrok | ok | 5.53s | 9,564 | 384 | 17 | 9,965 |
| speedclaude/haiku | ok | 9.08s | 272 | 0 | 559 | 831 |
| speedclaude/sonnet | ok | 7.44s | 322 | 0 | 265 | 587 |
| speedclaude/opus | ok | 6.86s | 306 | 0 | 241 | 547 |
| speedclaude/fable | ok | 7.58s | 307 | 0 | 241 | 548 |
| speedcodex/luna | ok | 14.0s | 12.2k | 0 | 255 | 12.4k |
| speedcodex/terra | ok | 10.6s | 5,437 | 6,912 | 243 | 12.6k |
| speedcodex/sol | ok | 9.26s | 12.3k | 0 | 243 | 12.6k |
| speedcodex/gpt-5.5 | ok | 10.4s | 11.4k | 0 | 257 | 11.7k |
| speedgrok/grok-4.6 | ok | 7.42s | 12.2k | 256 | 293 | 12.7k |
| speedgrok/grok-4.5 | ok | 7.86s | 12.3k | 128 | 280 | 12.7k |
Each row: start time · kind · trigger (cron means the timer, anything else means by hand) · wall time · rec / skip / fail · total tokens. Skipped means it was not that target's turn, or a rate limit stopped the app. Failed is a real error, and we keep it.
Sample log
Every measurement we saved, newest first. Filter by check or by app. Rows from old test kits stay here too — the boards ignore them, but we never delete them. The counts above always cover the full window, not the filter.
This window, counted by outcome
- ok · 52
- rate limit warning · 10
Every saved sample in the window, sorted into its label. The filters below change the table, never these counts.
The status words
- ok
- The app answered and the answer passed our grader.
- rate limit
- The vendor refused the request. We stop the rest of that app for the round. Busy plan, not a slow model.
- rate limit warning
- The vendor served the request but flagged that the plan's usage window is getting full. The run worked, but we keep it out of the medians to be safe. It never stops the round.
- timeout
- The app never finished in the time we allow.
- auth
- The sign-in was not accepted. We skip that app until it is fixed.
- used tools
- The app reached for a tool even though we said not to. The row does not count.
- wrong answer
- Text came back, but it was not what we asked for. Speed with a wrong answer never counts.
- remapped
- We asked for one model and the app served another. We save both names and mark the row, and the boards treat it with care.
- parse
- The app’s own output was not readable, so we could not trust the timings.
- overloaded / credits / failed
- The vendor said it was busy, out of credit, or gave a plain error.
| Started | Check | Target | Status | Wall | Total | Cost |
|---|---|---|---|---|---|---|
| Aug 21, 12:01:39 AM | speed | grok/grok-4.5grok-4.5-build | ok | 6.09s | 12.7k | $0.0089 |
| Aug 21, 12:01:32 AM | speed | grok/grok-4.6grok-4.6-build | ok | 6.92s | 12.7k | $0.0044 |
| Aug 21, 12:01:22 AM | speed | codex/gpt-5.5 | ok | 9.63s | 11.6k | — |
| Aug 21, 12:01:13 AM | speed | codex/sol | ok | 9.49s | 12.6k | — |
| Aug 21, 12:01:03 AM | speed | codex/terra | ok | 9.05s | 12.6k | — |
| Aug 21, 12:00:53 AM | speed | codex/luna | ok | 10.8s | 11.0k | — |
| Aug 21, 12:00:45 AM | speed | claude/fableclaude-fable-5served with a rate limit warning: seven_day at 0.28 utilization | rate limit warning | 7.16s | 548 | $0.0161 |
| Aug 21, 12:00:38 AM | speed | claude/opusclaude-opus-5served with a rate limit warning: seven_day at 0.28 utilization | rate limit warning | 6.79s | 547 | $0.0085 |
| Aug 21, 12:00:32 AM | speed | claude/sonnetclaude-sonnet-5served with a rate limit warning: seven_day at 0.28 utilization | rate limit warning | 6.91s | 563 | $0.0056 |
| Aug 21, 12:00:25 AM | speed | claude/haikuclaude-haiku-4-5served with a rate limit warning: seven_day at 0.28 utilization | rate limit warning | 6.09s | 830 | $0.0041 |
| Aug 21, 12:00:19 AM | ping | grok/grok-4.6grok-4.6-build | ok | 6.14s | 9,965 | $0.0033 |
| Aug 21, 12:00:10 AM | ping | codex/luna | ok | 8.59s | 12.2k | — |
| Aug 21, 12:00:03 AM | ping | claude/haikuclaude-haiku-4-5served with a rate limit warning: seven_day at 0.28 utilization | rate limit warning | 6.91s | 380 | $0.0019 |
| Aug 20, 11:01:27 PM | speed | codex/gpt-5.5 | ok | 10.4s | 11.6k | — |
| Aug 20, 11:01:17 PM | speed | codex/sol | ok | 9.71s | 12.6k | — |
| Aug 20, 11:01:07 PM | speed | codex/terra | ok | 9.83s | 12.6k | — |
| Aug 20, 11:00:53 PM | speed | codex/luna | ok | 14.4s | 12.4k | — |
| Aug 20, 11:00:45 PM | speed | claude/fableclaude-fable-5served with a rate limit warning: seven_day at 0.26 utilization | rate limit warning | 7.68s | 548 | $0.0161 |
| Aug 20, 11:00:36 PM | speed | claude/opusclaude-opus-5served with a rate limit warning: seven_day at 0.26 utilization | rate limit warning | 8.12s | 547 | $0.0085 |
| Aug 20, 11:00:30 PM | speed | claude/sonnetclaude-sonnet-5served with a rate limit warning: seven_day at 0.26 utilization | rate limit warning | 6.57s | 563 | $0.0056 |
| Aug 20, 11:00:18 PM | speed | claude/haikuclaude-haiku-4-5served with a rate limit warning: seven_day at 0.26 utilization | rate limit warning | 11.7s | 803 | $0.0039 |
| Aug 20, 11:00:09 PM | ping | codex/luna | ok | 9.19s | 12.2k | — |
| Aug 20, 11:00:04 PM | ping | claude/haikuclaude-haiku-4-5served with a rate limit warning: seven_day at 0.26 utilization | rate limit warning | 4.20s | 335 | $0.0016 |
| Aug 20, 10:02:00 PM | speed | grok/grok-4.5grok-4.5-build | ok | 8.33s | 12.7k | $0.0087 |
| Aug 20, 10:01:52 PM | speed | grok/grok-4.6grok-4.6-build | ok | 7.59s | 12.7k | $0.0045 |
| Aug 20, 10:01:38 PM | speed | codex/gpt-5.5 | ok | 12.9s | 13.0k | — |
| Aug 20, 10:01:22 PM | speed | codex/sol | ok | 15.8s | 12.6k | — |
| Aug 20, 10:01:10 PM | speed | codex/terra | ok | 12.4s | 12.6k | — |
| Aug 20, 10:00:55 PM | speed | codex/luna | ok | 14.9s | 12.4k | — |
| Aug 20, 10:00:47 PM | speed | claude/fableclaude-fable-5 | ok | 8.41s | 548 | $0.0161 |
| Aug 20, 10:00:39 PM | speed | claude/opusclaude-opus-5 | ok | 7.45s | 547 | $0.0085 |
| Aug 20, 10:00:32 PM | speed | claude/sonnetclaude-sonnet-5 | ok | 7.47s | 563 | $0.0056 |
| Aug 20, 10:00:21 PM | speed | claude/haikuclaude-haiku-4-5 | ok | 10.3s | 824 | $0.0040 |
| Aug 20, 10:00:16 PM | ping | grok/grok-4.6grok-4.6-build | ok | 5.24s | 9,978 | $0.0034 |
| Aug 20, 10:00:07 PM | ping | codex/luna | ok | 8.67s | 12.2k | — |
| Aug 20, 10:00:02 PM | ping | claude/haikuclaude-haiku-4-5 | ok | 4.56s | 294 | $0.0014 |
| Aug 20, 9:01:47 PM | speed | grok/grok-4.5grok-4.5-build | ok | 6.68s | 12.7k | $0.0089 |
| Aug 20, 9:01:40 PM | speed | grok/grok-4.6grok-4.6-build | ok | 7.20s | 12.7k | $0.0045 |
| Aug 20, 9:01:31 PM | speed | codex/gpt-5.5 | ok | 8.86s | 11.5k | — |
| Aug 20, 9:01:20 PM | speed | codex/sol | ok | 10.9s | 12.6k | — |
| Aug 20, 9:01:08 PM | speed | codex/terra | ok | 11.6s | 12.6k | — |
| Aug 20, 9:00:55 PM | speed | codex/luna | ok | 13.7s | 12.4k | — |
| Aug 20, 9:00:46 PM | speed | claude/fableclaude-fable-5 | ok | 7.90s | 548 | $0.0161 |
| Aug 20, 9:00:38 PM | speed | claude/opusclaude-opus-5 | ok | 8.38s | 547 | $0.0085 |
| Aug 20, 9:00:31 PM | speed | claude/sonnetclaude-sonnet-5 | ok | 7.17s | 583 | $0.0059 |
| Aug 20, 9:00:22 PM | speed | claude/haikuclaude-haiku-4-5 | ok | 8.82s | 817 | $0.0040 |
| Aug 20, 9:00:15 PM | ping | grok/grok-4.6grok-4.6-build | ok | 6.50s | 9,978 | $0.0033 |
| Aug 20, 9:00:06 PM | ping | codex/luna | ok | 8.51s | 12.2k | — |
| Aug 20, 9:00:02 PM | ping | claude/haikuclaude-haiku-4-5 | ok | 4.62s | 312 | $0.0015 |
| Aug 20, 8:01:45 PM | speed | grok/grok-4.5grok-4.5-build | ok | 7.86s | 12.7k | $0.0089 |
| Aug 20, 8:01:38 PM | speed | grok/grok-4.6grok-4.6-build | ok | 7.42s | 12.7k | $0.0045 |
| Aug 20, 8:01:27 PM | speed | codex/gpt-5.5 | ok | 10.4s | 11.7k | — |
| Aug 20, 8:01:18 PM | speed | codex/sol | ok | 9.26s | 12.6k | — |
| Aug 20, 8:01:07 PM | speed | codex/terra | ok | 10.6s | 12.6k | — |
| Aug 20, 8:00:53 PM | speed | codex/luna | ok | 14.0s | 12.4k | — |
| Aug 20, 8:00:46 PM | speed | claude/fableclaude-fable-5 | ok | 7.58s | 548 | $0.0161 |
| Aug 20, 8:00:39 PM | speed | claude/opusclaude-opus-5 | ok | 6.86s | 547 | $0.0085 |
| Aug 20, 8:00:31 PM | speed | claude/sonnetclaude-sonnet-5 | ok | 7.44s | 587 | $0.0059 |
| Aug 20, 8:00:22 PM | speed | claude/haikuclaude-haiku-4-5 | ok | 9.08s | 831 | $0.0040 |
| Aug 20, 8:00:16 PM | ping | grok/grok-4.6grok-4.6-build | ok | 5.53s | 9,965 | $0.0033 |
| Aug 20, 8:00:07 PM | ping | codex/luna | ok | 8.96s | 12.2k | — |
| Aug 20, 8:00:02 PM | ping | claude/haikuclaude-haiku-4-5 | ok | 4.79s | 293 | $0.0014 |
62 samples in the window. Same rows as JSON: /api/probes · /api/speed · /api/runs
Read next
- How we test
What each check asks, and what every unit means.
- Lab health
Is the lab fresh right now, and is it still signed in?
- Developers
The same data as JSON. No key, no login.