Model monitor
Moonshot Kimi K2.5
moonshotai/kimi-k2.5 · via openrouter
As of 2026-08-05: no alarm in 5 of 5 task families · 103 complete panels behind the record.
Served route, trailing 30 days: median latency 36.5 s · p95 128.1 s across 834 probe calls · served by DigitalOcean (819), unrecorded (326), DeepInfra (15). 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 15:06 UTC | 1.000.00 rate limited0.00 rate limited0.00 truncated1.00 | 0.400 | b3cda9518b92… |
| 44 | 2026-08-05 14:04 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 15d9e1cd0cc8… |
| 43 | 2026-08-05 13:03 UTC | 1.001.001.000.00 timeout1.00 | 0.800 | d5125e070096… |
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
4bf2bb89-7823-44ce-98ec-289f3fdc20c8- Opened
- 2026-08-05 12:27 UTC
- Previous epoch
9ace9483-a2c0-48b1-8a4f-92c4d2f72d23- 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
- -0.11 · max 0.00 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 10
- Provider spend, this epoch
- $0.0468778
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 | 0.00 rate limited0.00 truncated0.00 rate limited0.001.00 | 0.200 | 5d84a20ec7c1… |
| 44 | 2026-08-05 14:00 UTC | 1.001.001.000.001.00 | 0.800 | 1a8e68801c74… |
| 43 | 2026-08-05 12:58 UTC | 1.001.001.000.001.00 | 0.800 | 666ee8bc753f… |
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
a6d5277d-57c2-4d96-8d57-7ff3c259283c- Opened
- 2026-08-05 12:27 UTC
- Previous epoch
851d9d2d-1485-4dfd-9184-2b041bc0ff82- 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
- 12
- Provider spend, this epoch
- $0.0300733
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
- 47
- CS lower
- 0.503
- CS upper
- 0.864
Panel record — 47 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 46 | 2026-08-05 15:48 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | 7a509dd43324… |
| 45 | 2026-08-05 14:45 UTC | 1.001.000.00 truncated1.000.00 rate limited | 0.600 | c93e14960af4… |
| 44 | 2026-08-05 13:41 UTC | 1.001.001.001.000.83 | 0.967 | 034abaa318aa… |
| 43 | 2026-08-05 12:39 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | f87a5a73f91a… |
| 42 | 2026-08-05 11:37 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | aec1c69f7980… |
| 41 | 2026-08-05 10:33 UTC | 1.001.000.831.000.83 | 0.933 | 958d2171e90d… |
| 40 | 2026-08-05 09:31 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | ff54f33fee16… |
| 39 | 2026-08-05 08:28 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | e6974fff67ce… |
| 38 | 2026-08-05 07:22 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 00911babaf79… |
| 37 | 2026-08-05 06:23 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 6a8258257324… |
| 36 | 2026-08-05 05:22 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 73d0e06922a2… |
| 35 | 2026-08-05 04:18 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 350c02f856e8… |
| 34 | 2026-08-05 03:17 UTC | 1.001.001.001.000.83 | 0.967 | e7cfbdde7168… |
| 33 | 2026-08-05 02:15 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.83 | 0.167 | 5470c92dddd7… |
| 32 | 2026-08-05 01:18 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 0bc9e5153cf2… |
| 31 | 2026-08-05 00:11 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 2168beb974a5… |
| 30 | 2026-08-04 23:12 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 2630117ff4b5… |
| 29 | 2026-08-04 22:12 UTC | 0.00 rate limited0.00 rate limited0.00 truncated0.00 rate limited0.83 | 0.167 | ea724bd2c488… |
| 28 | 2026-08-04 21:11 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | afda09712298… |
| 27 | 2026-08-04 20:10 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | d57edccabd69… |
| 26 | 2026-08-04 19:06 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 76e9cac35ee1… |
| 25 | 2026-08-04 18:03 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | d22dfd661008… |
| 24 | 2026-08-04 16:58 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 8034ab84968d… |
| 23 | 2026-08-04 15:54 UTC | 1.001.000.00 truncated0.00 truncated0.83 | 0.567 | e976797b716a… |
| 22 | 2026-08-04 14:52 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 6f509c8876e0… |
| 21 | 2026-08-04 13:45 UTC | 1.001.000.00 provider error1.000.83 | 0.767 | 8d01ac755d8e… |
| 20 | 2026-08-04 12:43 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 16e3fa5ae391… |
| 19 | 2026-08-04 11:45 UTC | 1.001.000.00 timeout1.000.83 | 0.767 | 7c749d9c379b… |
| 18 | 2026-08-04 10:42 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | 6bd0e23d3bff… |
| 17 | 2026-08-04 09:39 UTC | 1.001.000.00 truncated1.000.83 | 0.767 | ac011516f784… |
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
91d10d90-5b83-4101-8aea-35f16ee28a75- Opened
- 2026-08-03 16:27 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
- -0.37 · max 1.02 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 166
- Provider spend, this epoch
- $0.5290036
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 15:37 UTC | 0.00 truncated0.00 truncated0.00 rate limited0.00 truncated0.00 rate limited | 0.000 | d0cc38957319… |
| 44 | 2026-08-05 14:34 UTC | 0.00 truncated1.001.001.000.00 truncated | 0.600 | 79408b8cfb4b… |
| 43 | 2026-08-05 13:33 UTC | 0.00 truncated1.001.001.000.00 truncated | 0.600 | f57c5118a3ab… |
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
fbac9ea0-9c5d-48d1-8454-df1049493db7- Opened
- 2026-08-05 12:27 UTC
- Previous epoch
2582d56c-7eab-4b5e-b1d1-1314be891eb3- 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
- 0.81 · max 0.81 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 6
- Provider spend, this epoch
- $0.0626554
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
- 47
- CS lower
- 0.584
- CS upper
- 0.922
Panel record — 47 completed panels
| Cycle | Completed | Member scores | Mean | Evidence |
|---|---|---|---|---|
| 46 | 2026-08-05 15:45 UTC | 1.001.001.000.00 rate limited1.00 | 0.800 | 3434ebf64857… |
| 45 | 2026-08-05 14:38 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 347fe026cf38… |
| 44 | 2026-08-05 13:37 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 0f0318e1d2aa… |
| 43 | 2026-08-05 12:35 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 76b28106a923… |
| 42 | 2026-08-05 11:35 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | e4d26b04628c… |
| 41 | 2026-08-05 10:29 UTC | 0.00 provider error1.000.00 provider error0.00 provider error1.00 | 0.400 | 2451bbc20728… |
| 40 | 2026-08-05 09:24 UTC | 0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited0.00 rate limited | 0.000 | a3413eb82d3b… |
| 39 | 2026-08-05 08:19 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 9b5d9a083de9… |
| 38 | 2026-08-05 07:18 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | e2ac1795bbb2… |
| 37 | 2026-08-05 06:11 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 5345566d9301… |
| 36 | 2026-08-05 05:10 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | bd2a33621fa9… |
| 35 | 2026-08-05 04:07 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 1ccdeeebdc4d… |
| 34 | 2026-08-05 03:10 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 915942c47ebb… |
| 33 | 2026-08-05 02:05 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 9b9ae9fa9017… |
| 32 | 2026-08-05 01:04 UTC | 1.001.001.000.00 timeout0.00 rate limited | 0.600 | 01d3f8b3eedc… |
| 31 | 2026-08-04 23:57 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 9e2ba90e53f1… |
| 30 | 2026-08-04 22:59 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | b4a06675d5da… |
| 29 | 2026-08-04 21:58 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 68e07d67fa28… |
| 28 | 2026-08-04 20:57 UTC | 1.001.001.000.00 timeout1.00 | 0.800 | 06a8c496dc61… |
| 27 | 2026-08-04 19:56 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | a425f0af8fdf… |
| 26 | 2026-08-04 18:53 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | aa106da5bdc4… |
| 25 | 2026-08-04 17:49 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 4e512bef726d… |
| 24 | 2026-08-04 16:50 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 8a57d95aeb14… |
| 23 | 2026-08-04 15:47 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | b6946fdfba4a… |
| 22 | 2026-08-04 14:42 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 7488b3e8249f… |
| 21 | 2026-08-04 13:39 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 12c0c0920501… |
| 20 | 2026-08-04 12:40 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 1bcf931aee17… |
| 19 | 2026-08-04 11:38 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 933edaf8b132… |
| 18 | 2026-08-04 10:32 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 491a8731a72e… |
| 17 | 2026-08-04 09:30 UTC | 1.001.001.000.00 truncated1.00 | 0.800 | 3a94ae17f458… |
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
2a0bd841-48bc-4078-8a50-a0c83dfcf580- Opened
- 2026-08-03 16:27 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
- -0.49 · max 0.42 · alarm at 3.00 (= ln 1/α)
- Raw scores kept
- 177
- Provider spend, this epoch
- $0.4383039
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.