Actuary
← Moonshot Kimi K2.5

Evidence block

Structured output · cycle 439

moonshotai/kimi-k2.5 · 2026-08-22 15:49:48 UTC

This page is the full public record of one completed panel — the record’s hash-committed unit of evidence. Its identity is its hash; nothing here can change without changing the URL.

Identity

What this block is.

Panel mean
0.800
Members
5
Checker scored
4
Frozen floor
0.700
Anytime level α
0.050
Hashes and versions
Evidence hash
a681122c759c93d591b56b75948fedf66bf0aab28e7fa6a525007702d1241d98
Design hash
8a4f5431d708d308ccd859fc31d097a5db0297535ace2795cab39b5f087eb318
Panel hash
a37ad21aa4b8c3e16dbae958c68a70064a9d95487e131d832f2c601f7fdacf14
Stream
fbdf265a-e12a-4d13-855e-43333ee3312e
Epoch
4bf2bb89-7823-44ce-98ec-289f3fdc20c8
Pool version
structured-output-v3
Checker version
structured_output-v2
Panel size
5
Provider spend
$0.013801

Checker layer

Panel members.

# Item Item id Outcome Score
1 structured-output-015 390b56f6-0b7a-471d-a7dd-24e4a6519560 ok 1.00
2 structured-output-045 60032adc-9d89-4ecd-9431-4a3ecab9c9c6 ok 1.00
3 structured-output-075 b2f95fda-69c8-404c-8445-dcf17532fbee ok 1.00
4 structured-output-105 51363b3d-6e2e-4fd0-a633-3f6bcfa34fe9 truncated 0.00
5 structured-output-135 f1e6b049-9fe5-48da-b032-6034aefe87ca ok 1.00

Deterministic checkers score every response in [0, 1]; a member that ends in a terminal failure scores 0 and stays on the record. Item text stays unpublished so nobody can tune against the panel — identity travels as the item id and its pool source id.

Capture layer

Provider requests.

# Provider Upstream Response model Response id Finish Latency ms Tokens in Tokens out Cost Output SHA-256 Observed
1 openrouter DeepInfra moonshotai/kimi-k2.5 gen-1787413771-68vyl6JgX02A1igS9ypu stop 10112 79 1153 $0.0026298 ece676d37d4e634fd53719aa6df6ef9334f308d465cfd45034bb4e4fafa6f4d2 2026-08-22 15:49:41 UTC
2 openrouter DeepInfra moonshotai/kimi-k2.5 gen-1787413775-tp0yeC3rfc2SwYQMOkL2 stop 4754 84 398 $0.0009333 83f204ed0c10f7eafef22ab193c3a53dfa6f54808b8101ceab29beb23ed9a84e 2026-08-22 15:49:39 UTC
3 openrouter DeepInfra moonshotai/kimi-k2.5 gen-1787413769-H90d5YJ6hnT3S1ffQWCy stop 6644 110 876 $0.0020205 ab7ee00599637be7807315ff1e42f79a2d43b60ef436a9e7f71467b8e0b3721e 2026-08-22 15:49:36 UTC
4 openrouter — — — — — 195 2048 $0.0046958 — 2026-08-22 15:49:47 UTC
5 openrouter DeepInfra moonshotai/kimi-k2.5 gen-1787413776-DkudDSiLu2B96ZxiGgBc stop 12188 156 1534 $0.0035217 5c643f8dcbbdb8094bdeb75876e756c349a9c6574625437f717818913662c3bf 2026-08-22 15:49:48 UTC

Everything a probe captures, as captured. Raw model text is scrubbed thirty days after the run; its SHA-256, the outcome and the score are permanent. Costs are provider-reported and include reserved estimates where a run’s true cost was unrecoverable; a dash is a field the provider did not return.

Verify

Recompute the hash yourself.

The JSON twin of this page carries the exact evidence object this URL is named after. Canonicalise it per RFC 8785 and hash it; the digest must equal the address of this page.

GET /api/v1/evidence/a681122c759c…

curl -s https://actuary.sh/api/v1/evidence/a681122c759c93d591b56b75948fedf66bf0aab28e7fa6a525007702d1241d98 \
  | jq -cjS .evidence | shasum -a 256

For this object’s field types, jq -cjS reproduces RFC 8785 canonical form, and the JSON twin serves the hashed bytes verbatim. Blocks of exactly this shape appear verbatim in signed certificates — verify one offline, or read how the e-process turns panel means into an alarm.