Model monitor
Z.ai GLM 5.2
z-ai/glm-5.2 · via openrouter
As of 2026-08-05: no alarm in 5 of 5 task families · 91 complete panels behind the record.
Served route, trailing 30 days: median latency 6.7 s · p95 22.4 s across 918 probe calls · served by Decart (510), StreamLake (249), Novita (159), unrecorded (92). Observed at panel time — a few calls a day, not continuous uptime.
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
- 3
- CS lower
- 0.000
- CS upper
- 1.000
Panel record — 3 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 45 | 2026-08-05 15:05 UTC | 1.001.001.000.381.00 | 0.875 | a407d6d1221e… |
| 44 | 2026-08-05 14:01 UTC | 1.001.001.001.001.00 | 1.000 | 5d5a6c1625ab… |
| 43 | 2026-08-05 12:59 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 12cf02d36e4c… |
Newest first. 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
fc8c2825-f6d1-4156-ae09-e4b4540801e6- Opened
- 2026-08-05 12:27 UTC
- Previous epoch
41fd71dc-651e-4961-8a42-bab6bc083093- 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
- -0.19 · max 0.00 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 14
- Provider spend, this epoch
- $0.012992
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
- 3
- CS lower
- 0.000
- CS upper
- 1.000
Panel record — 3 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 45 | 2026-08-05 15:04 UTC | 1.001.001.000.001.00 | 0.800 | 37e88141536e… |
| 44 | 2026-08-05 13:59 UTC | 1.001.001.000.001.00 | 0.800 | 2ef22c3674ff… |
| 43 | 2026-08-05 12:57 UTC | 1.001.001.000.001.00 | 0.800 | e85f35952f61… |
Newest first. 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
729f1f2e-ee20-4455-8d24-6be9670d68d3- Opened
- 2026-08-05 12:54 UTC
- Previous epoch
220d13e9-5c31-43ad-a7e8-2ba2a5d7b270- 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
- -0.11 · max 0.00 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 15
- Provider spend, this epoch
- $0.0069001
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
- 41
- CS lower
- 0.667
- CS upper
- 1.000
Panel record — 41 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 46 | 2026-08-05 15:49 UTC | 1.001.001.001.001.00 | 1.000 | 3a9f822747c6… |
| 45 | 2026-08-05 14:43 UTC | 1.001.000.00 truncated1.001.00 | 0.800 | 94a4f695cc32… |
| 44 | 2026-08-05 13:39 UTC | 1.001.001.001.001.00 | 1.000 | e7a88a588481… |
| 43 | 2026-08-05 12:37 UTC | 1.001.000.00 truncated1.001.00 | 0.800 | 98e45223a253… |
| 42 | 2026-08-05 11:36 UTC | 1.001.001.001.001.00 | 1.000 | 23173431cbf5… |
| 41 | 2026-08-05 10:32 UTC | 1.001.000.00 truncated1.001.00 | 0.800 | 0598ccb3f34c… |
| 40 | 2026-08-05 09:31 UTC | 1.001.001.001.001.00 | 1.000 | b0c3fc4715a2… |
| 39 | 2026-08-05 08:27 UTC | 1.001.001.001.001.00 | 1.000 | e953d5484670… |
| 38 | 2026-08-05 07:22 UTC | 1.001.001.001.001.00 | 1.000 | 4c0a5b196162… |
| 37 | 2026-08-05 06:23 UTC | 1.001.000.00 truncated1.001.00 | 0.800 | dafa28bebb21… |
| 36 | 2026-08-05 05:22 UTC | 1.001.001.001.001.00 | 1.000 | ca4b5c0fe08f… |
| 35 | 2026-08-05 04:18 UTC | 1.001.001.001.001.00 | 1.000 | ad0ff2b01cd8… |
| 34 | 2026-08-05 03:17 UTC | 1.001.000.00 truncated1.001.00 | 0.800 | d3c6f727ba51… |
| 33 | 2026-08-05 02:15 UTC | 1.001.001.001.001.00 | 1.000 | 833129bee17e… |
| 32 | 2026-08-05 01:15 UTC | 1.001.001.001.001.00 | 1.000 | fe9c4b03f2af… |
| 31 | 2026-08-05 00:10 UTC | 1.001.000.00 truncated1.001.00 | 0.800 | 6d7fcf021cef… |
| 30 | 2026-08-04 23:11 UTC | 1.001.000.00 truncated1.001.00 | 0.800 | c447ac047174… |
| 29 | 2026-08-04 22:09 UTC | 1.001.001.001.001.00 | 1.000 | 181b25dfaa8e… |
| 28 | 2026-08-04 21:08 UTC | 1.001.001.001.001.00 | 1.000 | 0237dbca5cb1… |
| 27 | 2026-08-04 20:08 UTC | 1.001.001.001.001.00 | 1.000 | d0e34f17666c… |
| 26 | 2026-08-04 19:04 UTC | 1.001.001.001.001.00 | 1.000 | 5789c802eebb… |
| 25 | 2026-08-04 18:01 UTC | 1.001.001.001.001.00 | 1.000 | f127d9f53f78… |
| 24 | 2026-08-04 16:57 UTC | 1.001.000.00 truncated1.001.00 | 0.800 | db80b5bbf59a… |
| 23 | 2026-08-04 15:54 UTC | 1.001.001.001.001.00 | 1.000 | 2e7bb4d342dc… |
| 22 | 2026-08-04 14:49 UTC | 1.001.001.001.001.00 | 1.000 | 59dddfa4273e… |
| 21 | 2026-08-04 13:45 UTC | 0.00 rate limited1.001.001.001.00 | 0.800 | 7a747b33a92a… |
| 20 | 2026-08-04 12:42 UTC | 1.000.00 rate limited0.00 rate limited1.000.00 rate limited | 0.400 | 61aef268b9fa… |
| 19 | 2026-08-04 11:42 UTC | 1.000.00 rate limited0.00 truncated1.000.00 rate limited | 0.400 | b4da17238c06… |
| 18 | 2026-08-04 10:41 UTC | 1.001.000.00 rate limited1.001.00 | 0.800 | b6fd5aadbc00… |
| 17 | 2026-08-04 09:37 UTC | 1.001.000.00 truncated1.001.00 | 0.800 | 62e0fce2142f… |
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
37cb9150-9ea9-4f2f-baa9-815ec952f905- Opened
- 2026-08-03 21:54 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
- -0.51 · max 0.39 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 178
- Provider spend, this epoch
- $0.2724834
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
- Frozen null floor
- 0.700
- Anytime level α
- 0.050
- Completed panels
- 3
- CS lower
- 0.000
- CS upper
- 1.000
Panel record — 3 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 45 | 2026-08-05 15:33 UTC | 1.001.001.001.001.00 | 1.000 | e91c5c2fb1d2… |
| 44 | 2026-08-05 14:32 UTC | 1.001.001.001.001.00 | 1.000 | d95ce59b6c27… |
| 43 | 2026-08-05 13:30 UTC | 1.001.001.001.001.00 | 1.000 | 0658b73f919c… |
Newest first. 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
67c4ddac-b64d-4463-8a2e-a79858ed55b2- Opened
- 2026-08-05 12:27 UTC
- Previous epoch
729f6eb4-ad1a-40c5-80ec-09aebbcd86b8- 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
- -0.23 · max 0.00 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 15
- Provider spend, this epoch
- $0.0077757
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
- 41
- CS lower
- 0.565
- CS upper
- 0.966
Panel record — 41 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 46 | 2026-08-05 15:44 UTC | 0.001.001.001.001.00 | 0.800 | d8c6bbe555cf… |
| 45 | 2026-08-05 14:34 UTC | 0.001.001.000.001.00 | 0.600 | 0c7156cd086d… |
| 44 | 2026-08-05 13:34 UTC | 0.001.001.001.001.00 | 0.800 | 1886accb49fd… |
| 43 | 2026-08-05 12:32 UTC | 1.001.001.001.001.00 | 1.000 | 4da2ea2771a0… |
| 42 | 2026-08-05 11:33 UTC | 0.001.001.001.001.00 | 0.800 | 40e2317e5b27… |
| 41 | 2026-08-05 10:28 UTC | 0.001.001.000.001.00 | 0.600 | 6136afcff9eb… |
| 40 | 2026-08-05 09:24 UTC | 0.001.001.001.001.00 | 0.800 | 4007c3b0f1fd… |
| 39 | 2026-08-05 08:18 UTC | 0.001.001.001.001.00 | 0.800 | 357ec63d591d… |
| 38 | 2026-08-05 07:17 UTC | 1.001.001.001.001.00 | 1.000 | 20cc6fc94a98… |
| 37 | 2026-08-05 06:10 UTC | 1.001.001.001.001.00 | 1.000 | f419729369a8… |
| 36 | 2026-08-05 05:09 UTC | 1.001.001.001.001.00 | 1.000 | 72e905987372… |
| 35 | 2026-08-05 04:06 UTC | 0.001.001.001.001.00 | 0.800 | fbcf71f61dcb… |
| 34 | 2026-08-05 03:09 UTC | 0.001.001.001.001.00 | 0.800 | 8aec6008ca5a… |
| 33 | 2026-08-05 02:05 UTC | 1.001.001.001.001.00 | 1.000 | 0e6efe098977… |
| 32 | 2026-08-05 01:01 UTC | 1.001.001.000.001.00 | 0.800 | f09b2b9afd76… |
| 31 | 2026-08-04 23:56 UTC | 0.001.001.001.001.00 | 0.800 | 90a090522460… |
| 30 | 2026-08-04 22:57 UTC | 0.001.001.000.001.00 | 0.600 | c723cd9d339d… |
| 29 | 2026-08-04 21:55 UTC | 0.001.001.001.001.00 | 0.800 | 91ea3f61f372… |
| 28 | 2026-08-04 20:54 UTC | 0.001.001.001.001.00 | 0.800 | dcae93b73171… |
| 27 | 2026-08-04 19:54 UTC | 0.001.001.001.001.00 | 0.800 | ce9c6ed17023… |
| 26 | 2026-08-04 18:51 UTC | 0.00 rate limited1.000.00 rate limited0.00 rate limited1.00 | 0.400 | de9dc161ff96… |
| 25 | 2026-08-04 17:47 UTC | 0.001.001.000.001.00 | 0.600 | 8ff0d7caede7… |
| 24 | 2026-08-04 16:48 UTC | 0.001.001.001.001.00 | 0.800 | 034cd638b392… |
| 23 | 2026-08-04 15:45 UTC | 0.001.001.001.001.00 | 0.800 | 6bc551c58f5a… |
| 22 | 2026-08-04 14:41 UTC | 0.001.001.000.001.00 | 0.600 | 077cf4a4d6ae… |
| 21 | 2026-08-04 13:39 UTC | 1.001.001.000.001.00 | 0.800 | 1b44f0c497e1… |
| 20 | 2026-08-04 12:38 UTC | 0.000.00 rate limited0.00 rate limited1.001.00 | 0.400 | ac35d5bccf13… |
| 19 | 2026-08-04 11:37 UTC | 0.000.00 rate limited1.001.000.00 rate limited | 0.400 | f8b732ee098e… |
| 18 | 2026-08-04 10:32 UTC | 1.001.001.001.001.00 | 1.000 | 483f694aaf2d… |
| 17 | 2026-08-04 09:28 UTC | 0.001.001.001.001.00 | 0.800 | c5e0ec785408… |
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
f070d7b2-7a0c-401f-9433-b25347aec444- Opened
- 2026-08-03 21:54 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
- -0.83 · max 0.20 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 189
- Provider spend, this epoch
- $0.1653484
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.