Model monitor
Qwen 3 30B A3B Instruct Reference
qwen/qwen3-30b-a3b-instruct-2507 · via openrouter
As of 2026-08-23: reference route · 1688 complete panels attest measurement across 5 task families.
Reference route: this model is the co-probed confound control. It never opens a public incident — its record attests measurement, not verdicts.
Certified claim, per family: H0: E[panel_mean_t | prior completed panels] ≥ frozen_null_floor. Fixed panel composition does not assume or establish temporal independence. Recent panels below link to their hash-committed evidence.
Task family
Structured output
- Frozen null floor
- 0.700
- Anytime level α
- 0.050
- Completed panels
- 418
- CS lower
- 0.903
- CS upper
- 0.978
Limited history: the chart is capped; current counts and confidence bounds are not.
Panel record — 418 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 460 | 2026-08-23 13:46 UTC | 1.001.001.001.001.00 | 1.000 | f65f3c1c2d7e… |
| 459 | 2026-08-23 12:45 UTC | 1.001.001.001.001.00 | 1.000 | 8012e19fc40a… |
| 458 | 2026-08-23 11:48 UTC | 1.001.001.001.001.00 | 1.000 | cf0d416b1d4e… |
| 457 | 2026-08-23 10:44 UTC | 1.001.001.001.001.00 | 1.000 | bac61cb0d9c6… |
| 456 | 2026-08-23 09:37 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | f060a2c0a65f… |
| 455 | 2026-08-23 08:37 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | 81f18ab3a746… |
| 454 | 2026-08-23 07:38 UTC | 1.001.001.001.001.00 | 1.000 | d16c2bfb9bf5… |
| 453 | 2026-08-23 06:33 UTC | 1.001.001.001.001.00 | 1.000 | 79637b323304… |
| 452 | 2026-08-23 05:33 UTC | 1.001.001.001.001.00 | 1.000 | 8b3a01833e47… |
| 451 | 2026-08-23 04:32 UTC | 1.001.001.001.001.00 | 1.000 | 5f43206b81ae… |
| 450 | 2026-08-23 03:24 UTC | 1.001.001.001.001.00 | 1.000 | 09d4a92d2d1d… |
| 449 | 2026-08-23 02:24 UTC | 1.001.001.001.001.00 | 1.000 | fff8fe86e06e… |
| 448 | 2026-08-23 01:24 UTC | 1.001.001.001.001.00 | 1.000 | 4b59b60a7e4f… |
| 447 | 2026-08-23 00:16 UTC | 1.001.001.001.001.00 | 1.000 | 1b097e2eb87b… |
| 446 | 2026-08-22 23:15 UTC | 1.001.001.001.001.00 | 1.000 | fc9adbe7fdef… |
| 445 | 2026-08-22 22:09 UTC | 1.001.001.001.001.00 | 1.000 | 4c21275e1366… |
| 444 | 2026-08-22 21:04 UTC | 1.001.001.001.001.00 | 1.000 | c1eaa0beacff… |
| 443 | 2026-08-22 20:01 UTC | 1.001.001.001.001.00 | 1.000 | 297bb045137f… |
| 442 | 2026-08-22 18:59 UTC | 1.001.001.001.001.00 | 1.000 | 171bed42d665… |
| 441 | 2026-08-22 17:54 UTC | 1.001.001.001.001.00 | 1.000 | 0245204f0a0a… |
| 440 | 2026-08-22 16:51 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | a876095075d4… |
| 439 | 2026-08-22 15:49 UTC | 1.001.001.001.001.00 | 1.000 | 7207b261e9d0… |
| 438 | 2026-08-22 14:49 UTC | 1.001.001.001.001.00 | 1.000 | 873cc94848de… |
| 437 | 2026-08-22 13:48 UTC | 1.001.001.001.001.00 | 1.000 | 3b9c315600c1… |
| 436 | 2026-08-22 12:47 UTC | 1.001.001.001.001.00 | 1.000 | 7fd40d7baa0c… |
| 435 | 2026-08-22 11:47 UTC | 1.001.001.001.001.00 | 1.000 | 5191172b2a6f… |
| 434 | 2026-08-22 10:41 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | ec323182b524… |
| 433 | 2026-08-22 09:41 UTC | 1.001.001.001.001.00 | 1.000 | 0d7f6023105c… |
| 432 | 2026-08-22 08:42 UTC | 1.001.001.001.001.00 | 1.000 | a36885902bae… |
| 431 | 2026-08-22 07:36 UTC | 1.001.001.001.001.00 | 1.000 | 49a11ea6ada7… |
Newest first; older panels are retained and replayable even where this table truncates. A failed member scores 0 and stays on the record. Each evidence link opens the panel’s full hash-committed record.
Epoch & design record
- Current epoch
cae9b52b-1f14-4aa4-a50f-c33a6b4fa544- Opened
- 2026-08-05 12:27 UTC
- Previous epoch
55048047-8796-48c4-bded-7f7538ceb6c9- Transition
- checker_refresh
- Design hash
8a4f5431d708d308ccd859fc31d097a5db0297535ace2795cab39b5f087eb318- Panel hash
a37ad21aa4b8c3e16dbae958c68a70064a9d95487e131d832f2c601f7fdacf14- Pool hash
8e9496011afbb949e38b3facc664e6dd560f1c5d2155fd424fcb720d084c69ee- Panel size
- 5
- Terminal failure
- zero — a failed member scores 0 and stays on the record
- Pool version
structured-output-v3- Checker version
structured_output-v2- Reference mean μ₀
- 0.700
- Minimum effect δ
- 0.000
- Evidence log e
- -0.23 · max 0.00 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 1957
- Provider spend, this epoch
- $0.1244956
The persisted monitor state is the authority a certificate replays byte-for-byte; the chart band above is a faithful replay of the same recorded panel means. Spend sums completed probe runs and includes reserved estimates where a run’s true cost was unrecoverable.
Task family
Extraction
- Frozen null floor
- 0.700
- Anytime level α
- 0.050
- Completed panels
- 418
- CS lower
- 0.652
- CS upper
- 0.708
Limited history: the chart is capped; current counts and confidence bounds are not.
Panel record — 418 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 460 | 2026-08-23 13:43 UTC | 1.001.001.000.000.67 | 0.733 | 153db6105d58… |
| 459 | 2026-08-23 12:43 UTC | 1.001.001.000.000.67 | 0.733 | 6e937e634444… |
| 458 | 2026-08-23 11:42 UTC | 1.001.001.000.000.67 | 0.733 | fc5f885b214f… |
| 457 | 2026-08-23 10:41 UTC | 1.001.001.000.000.67 | 0.733 | ddedf15d713c… |
| 456 | 2026-08-23 09:35 UTC | 1.001.001.000.000.67 | 0.733 | 6644f06ed658… |
| 455 | 2026-08-23 08:35 UTC | 1.001.001.000.000.67 | 0.733 | 020a63b7ea96… |
| 454 | 2026-08-23 07:33 UTC | 1.001.001.000.000.67 | 0.733 | 6f1867ca4177… |
| 453 | 2026-08-23 06:31 UTC | 1.001.001.000.000.67 | 0.733 | 69127bca6e89… |
| 452 | 2026-08-23 05:30 UTC | 1.001.001.000.000.67 | 0.733 | 2a21ea80ed9c… |
| 451 | 2026-08-23 04:28 UTC | 1.001.001.000.000.67 | 0.733 | 8e4320123bf7… |
| 450 | 2026-08-23 03:22 UTC | 1.001.001.000.000.00 rate limited | 0.600 | e3c35df22b22… |
| 449 | 2026-08-23 02:22 UTC | 1.001.001.000.000.67 | 0.733 | 3cae7d7dd8ea… |
| 448 | 2026-08-23 01:21 UTC | 1.001.001.000.000.67 | 0.733 | f73adfc61746… |
| 447 | 2026-08-23 00:15 UTC | 1.001.001.000.000.67 | 0.733 | cfe4de100504… |
| 446 | 2026-08-22 23:13 UTC | 1.001.001.000.000.67 | 0.733 | 4e66c907eab6… |
| 445 | 2026-08-22 22:08 UTC | 1.001.001.000.000.67 | 0.733 | 2ffbb53252ce… |
| 444 | 2026-08-22 21:03 UTC | 1.001.001.000.000.67 | 0.733 | c8478ee8d20f… |
| 443 | 2026-08-22 20:00 UTC | 1.001.001.000.000.67 | 0.733 | 0d903d464042… |
| 442 | 2026-08-22 18:58 UTC | 1.001.001.000.000.67 | 0.733 | 04f9c0affc1f… |
| 441 | 2026-08-22 17:53 UTC | 1.001.001.000.000.67 | 0.733 | 95ef55aae3e6… |
| 440 | 2026-08-22 16:48 UTC | 1.001.001.000.000.67 | 0.733 | 807d9511e005… |
| 439 | 2026-08-22 15:48 UTC | 1.001.001.000.000.67 | 0.733 | d7cb1c0812c8… |
| 438 | 2026-08-22 14:48 UTC | 1.001.001.000.000.67 | 0.733 | 53f428101c90… |
| 437 | 2026-08-22 13:47 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | 5ae05871a08b… |
| 436 | 2026-08-22 12:46 UTC | 1.001.001.000.000.67 | 0.733 | bb565d41ecba… |
| 435 | 2026-08-22 11:45 UTC | 1.001.001.000.000.67 | 0.733 | 29f40d60f3ef… |
| 434 | 2026-08-22 10:40 UTC | 1.001.001.000.000.67 | 0.733 | d8ceb0bb4257… |
| 433 | 2026-08-22 09:40 UTC | 1.001.001.000.000.67 | 0.733 | 1fee2cee34e4… |
| 432 | 2026-08-22 08:40 UTC | 1.001.001.000.000.67 | 0.733 | 68d1ab6f4ac9… |
| 431 | 2026-08-22 07:35 UTC | 1.001.001.000.000.67 | 0.733 | 4a2273afa06d… |
Newest first; older panels are retained and replayable even where this table truncates. A failed member scores 0 and stays on the record. Each evidence link opens the panel’s full hash-committed record.
Epoch & design record
- Current epoch
bf4e491f-636b-4982-b8e7-3ac9715fe38c- Opened
- 2026-08-05 12:27 UTC
- Previous epoch
21f69ed5-a596-4127-86cb-3d8dab24de2a- Transition
- checker_refresh
- Design hash
a2749265eac9731c94d1b8cada6fedc0e453bbc919053c455268b3a7eb248a61- Panel hash
1703e9221c9ce6ce1a9f69974cf0fd77056ac1ee4009dec47e3e6f23be2d4351- Pool hash
51802fc8964f7d9927bcabfae23b4bd2f0905e8977c9f8fb4ce50884ab0fbba6- Panel size
- 5
- Terminal failure
- zero — a failed member scores 0 and stays on the record
- Pool version
extraction-v3- Checker version
extraction-v2- Reference mean μ₀
- 0.700
- Minimum effect δ
- 0.000
- Evidence log e
- 3.50 · max 3.96 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 1917
- Provider spend, this epoch
- $0.0907993
The persisted monitor state is the authority a certificate replays byte-for-byte; the chart band above is a faithful replay of the same recorded panel means. Spend sums completed probe runs and includes reserved estimates where a run’s true cost was unrecoverable.
Task family
Instruction following
- Frozen null floor
- 0.700
- Anytime level α
- 0.050
- Completed panels
- 426
- CS lower
- 0.824
- CS upper
- 0.899
Limited history: the chart is capped; current counts and confidence bounds are not.
Panel record — 426 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 461 | 2026-08-23 13:44 UTC | 1.001.000.830.861.00 | 0.938 | f1c9a84ddbcc… |
| 460 | 2026-08-23 12:43 UTC | 1.001.000.830.861.00 | 0.938 | 5d9442e47059… |
| 459 | 2026-08-23 11:43 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | cc7ffa35eb7a… |
| 458 | 2026-08-23 10:41 UTC | 0.831.000.830.861.00 | 0.905 | 7aa007e1ff1a… |
| 457 | 2026-08-23 09:35 UTC | 0.831.000.830.861.00 | 0.905 | 8576c814a9e5… |
| 456 | 2026-08-23 08:35 UTC | 1.001.000.830.861.00 | 0.938 | 70da56282159… |
| 455 | 2026-08-23 07:34 UTC | 1.001.000.830.861.00 | 0.938 | 33198eb25979… |
| 454 | 2026-08-23 06:31 UTC | 0.831.000.830.861.00 | 0.905 | 9a89f669fa35… |
| 453 | 2026-08-23 05:31 UTC | 1.001.000.830.861.00 | 0.938 | cbeff44aa893… |
| 452 | 2026-08-23 04:29 UTC | 0.831.000.830.861.00 | 0.905 | 2f0772452c8c… |
| 451 | 2026-08-23 03:22 UTC | 0.831.000.830.861.00 | 0.905 | d02f00a96ce5… |
| 450 | 2026-08-23 02:22 UTC | 1.001.000.830.861.00 | 0.938 | a0762cb594c6… |
| 449 | 2026-08-23 01:22 UTC | 1.001.000.830.861.00 | 0.938 | d7b3222bdfea… |
| 448 | 2026-08-23 00:21 UTC | 1.001.000.830.861.00 | 0.938 | d95a62eb5267… |
| 447 | 2026-08-22 23:20 UTC | 1.001.000.830.861.00 | 0.938 | 575e0f4e4600… |
| 446 | 2026-08-22 22:14 UTC | 1.001.000.830.861.00 | 0.938 | 8aa445985496… |
| 445 | 2026-08-22 21:09 UTC | 1.001.000.830.861.00 | 0.938 | 67ba3cb13c96… |
| 444 | 2026-08-22 20:08 UTC | 0.831.000.830.861.00 | 0.905 | e98db8b40416… |
| 443 | 2026-08-22 19:06 UTC | 1.001.000.830.861.00 | 0.938 | ecb065f462f2… |
| 442 | 2026-08-22 18:00 UTC | 1.001.000.830.861.00 | 0.938 | f49151c2682f… |
| 441 | 2026-08-22 17:00 UTC | 1.001.000.830.861.00 | 0.938 | e8f20c25746b… |
| 440 | 2026-08-22 15:54 UTC | 0.831.000.830.861.00 | 0.905 | 388dba855a92… |
| 439 | 2026-08-22 14:55 UTC | 1.001.000.830.861.00 | 0.938 | f870edf094ad… |
| 438 | 2026-08-22 13:54 UTC | 1.001.000.830.861.00 | 0.938 | 19fdda147fe9… |
| 437 | 2026-08-22 12:53 UTC | 0.831.000.830.861.00 | 0.905 | c6a226d6f61e… |
| 436 | 2026-08-22 11:52 UTC | 0.831.000.830.861.00 | 0.905 | dac955191fdb… |
| 435 | 2026-08-22 10:46 UTC | 0.831.000.830.861.00 | 0.905 | 30b859463f81… |
| 434 | 2026-08-22 09:46 UTC | 1.001.000.830.861.00 | 0.938 | 03b0c1b6f2fe… |
| 433 | 2026-08-22 08:47 UTC | 1.001.000.830.861.00 | 0.938 | ec1f808b4cca… |
| 432 | 2026-08-22 07:41 UTC | 1.001.000.830.861.00 | 0.938 | dba4af2c67e7… |
Newest first; older panels are retained and replayable even where this table truncates. A failed member scores 0 and stays on the record. Each evidence link opens the panel’s full hash-committed record.
Epoch & design record
- Current epoch
2c5cd834-3470-4b44-bda8-0608c4762148- Opened
- 2026-08-05 04:17 UTC
- Design hash
4f5cc6b9ff4ce026ccab93f5a3c8241f78eb005bc8b8068e4c701e9236504412- Panel hash
e7dfea1f343f7bf37e066a1117ca21f3438e804ad61437bf72d810640cef7e71- Pool hash
9c5d065322394a6fb18c6d7f90c5bb63f00781f3e4c4aaa663536f76bb8ba7f9- Panel size
- 5
- Terminal failure
- zero — a failed member scores 0 and stays on the record
- Pool version
instruction_following-v3- Checker version
instruction_following-v1- Reference mean μ₀
- 0.700
- Minimum effect δ
- 0.000
- Evidence log e
- -0.18 · max 0.00 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 1955
- Provider spend, this epoch
- $0.0821008
The persisted monitor state is the authority a certificate replays byte-for-byte; the chart band above is a faithful replay of the same recorded panel means. Spend sums completed probe runs and includes reserved estimates where a run’s true cost was unrecoverable.
Task family
Code reasoning
Probing paused: nothing new lands on this family's record; everything already recorded below stands and stays citable.
- Frozen null floor
- 0.700
- Anytime level α
- 0.050
- Completed panels
- 0
This route is watched. Its record begins at the first complete panel — nothing is backfilled, including for us.
Epoch & design record
- Current epoch
05239de2-939a-40d1-b69d-49fc47c4ea60- Opened
- 2026-08-05 12:27 UTC
- Previous epoch
fd6173c2-1eef-48bd-acc8-d814c45c4766- Transition
- checker_refresh
- Design hash
bc3668bd52d9d48803b744c2c90f4ed816d7e394525b32b4219f6bab5c9c6df2- Panel hash
827de544c7641d140ac37e321bdb2578a41e291915f519183abbc5bd68153e81- Pool hash
cbfba66601ff51f43c83017d6de9e49988afcee72235f5dca89d705f4f8bb4cc- Panel size
- 5
- Terminal failure
- zero — a failed member scores 0 and stays on the record
- Pool version
code_reasoning-v4- Checker version
code_reasoning-v2- Reference mean μ₀
- 0.700
- Minimum effect δ
- 0.000
- Evidence log e
- 0.00 · max 0.00 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 0
The persisted monitor state is the authority a certificate replays byte-for-byte; the chart band above is a faithful replay of the same recorded panel means. Spend sums completed probe runs and includes reserved estimates where a run’s true cost was unrecoverable.
Task family
Math
- Frozen null floor
- 0.700
- Anytime level α
- 0.050
- Completed panels
- 426
- CS lower
- 0.878
- CS upper
- 0.965
Limited history: the chart is capped; current counts and confidence bounds are not.
Panel record — 426 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 461 | 2026-08-23 13:45 UTC | 1.001.001.001.001.00 | 1.000 | 829a865a4f10… |
| 460 | 2026-08-23 12:45 UTC | 1.001.001.001.001.00 | 1.000 | c58ca2091267… |
| 459 | 2026-08-23 11:45 UTC | 1.001.001.001.001.00 | 1.000 | 1b11ecbd6c08… |
| 458 | 2026-08-23 10:43 UTC | 1.001.001.001.001.00 | 1.000 | 0404c8e8168d… |
| 457 | 2026-08-23 09:36 UTC | 1.001.001.001.001.00 | 1.000 | 7c5da69edceb… |
| 456 | 2026-08-23 08:36 UTC | 1.001.001.001.001.00 | 1.000 | 3c000ba48aa6… |
| 455 | 2026-08-23 07:36 UTC | 1.001.001.001.001.00 | 1.000 | 94e040a60498… |
| 454 | 2026-08-23 06:32 UTC | 1.001.001.001.001.00 | 1.000 | 2c630d40983c… |
| 453 | 2026-08-23 05:32 UTC | 1.001.001.001.001.00 | 1.000 | f0b427870841… |
| 452 | 2026-08-23 04:31 UTC | 1.001.001.001.001.00 | 1.000 | 4cd874ab9fb4… |
| 451 | 2026-08-23 03:23 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | eab13fcf69c6… |
| 450 | 2026-08-23 02:23 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | c100cdcdaa87… |
| 449 | 2026-08-23 01:23 UTC | 1.001.001.001.001.00 | 1.000 | 5875f7aa5f4e… |
| 448 | 2026-08-23 00:22 UTC | 1.001.001.001.001.00 | 1.000 | 8e9b8cb2f258… |
| 447 | 2026-08-22 23:22 UTC | 1.001.001.001.001.00 | 1.000 | 7c3d137b4660… |
| 446 | 2026-08-22 22:15 UTC | 1.001.001.001.001.00 | 1.000 | df747ca44027… |
| 445 | 2026-08-22 21:11 UTC | 1.001.001.001.001.00 | 1.000 | aacc10a73a87… |
| 444 | 2026-08-22 20:12 UTC | 1.001.001.001.001.00 | 1.000 | 2277dbf716a7… |
| 443 | 2026-08-22 19:08 UTC | 1.001.001.001.001.00 | 1.000 | 7fee507a75d9… |
| 442 | 2026-08-22 18:01 UTC | 1.001.001.001.001.00 | 1.000 | 6fd21d3772fa… |
| 441 | 2026-08-22 17:01 UTC | 1.001.001.001.001.00 | 1.000 | cacc7f104b44… |
| 440 | 2026-08-22 15:55 UTC | 1.001.001.001.001.00 | 1.000 | 435cb457a639… |
| 439 | 2026-08-22 14:56 UTC | 1.001.001.001.001.00 | 1.000 | bc17639d6cc5… |
| 438 | 2026-08-22 13:56 UTC | 1.001.001.001.001.00 | 1.000 | 29ab3a740190… |
| 437 | 2026-08-22 12:56 UTC | 1.001.001.001.001.00 | 1.000 | a2749fe5ad3e… |
| 436 | 2026-08-22 11:54 UTC | 1.001.001.001.001.00 | 1.000 | 7e8caf678d5c… |
| 435 | 2026-08-22 10:47 UTC | 1.001.001.001.001.00 | 1.000 | 6ddd98447f75… |
| 434 | 2026-08-22 09:47 UTC | 1.001.001.001.001.00 | 1.000 | d8114747b233… |
| 433 | 2026-08-22 08:50 UTC | 1.001.001.001.001.00 | 1.000 | 484fe19f35f3… |
| 432 | 2026-08-22 07:42 UTC | 1.001.001.001.001.00 | 1.000 | b31424da54f8… |
Newest first; older panels are retained and replayable even where this table truncates. A failed member scores 0 and stays on the record. Each evidence link opens the panel’s full hash-committed record.
Epoch & design record
- Current epoch
f632e870-72ee-4ce5-bb51-fd0001dac3a3- Opened
- 2026-08-05 04:17 UTC
- Design hash
fdf94065c55c2ee7d86701efc50be0f5abdc5f02a236f6df52c2ac6f7db801eb- Panel hash
882f3932d07aeb85c4f6406a550d8ad14512e70d3df2432fe7aab03d0508bc75- Pool hash
7ed49fa9a63ba0e0907ff8326de00a843a926618f6d0dec40aa8e491d041e99b- Panel size
- 5
- Terminal failure
- zero — a failed member scores 0 and stays on the record
- Pool version
math-v4- Checker version
math-v1- Reference mean μ₀
- 0.700
- Minimum effect δ
- 0.000
- Evidence log e
- -0.23 · max 0.00 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 1961
- Provider spend, this epoch
- $0.1571313
The persisted monitor state is the authority a certificate replays byte-for-byte; the chart band above is a faithful replay of the same recorded panel means. Spend sums completed probe runs and includes reserved estimates where a run’s true cost was unrecoverable.