Actuary
← Moonshot Kimi K2.5

Evidence block

Code reasoning · cycle 458

moonshotai/kimi-k2.5 · 2026-08-23 11:42:21 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
36a4ce89e69db47b493609110db3d7735fcebf3bbd5d7ca9b533619de96d8929
Design hash
bc3668bd52d9d48803b744c2c90f4ed816d7e394525b32b4219f6bab5c9c6df2
Panel hash
827de544c7641d140ac37e321bdb2578a41e291915f519183abbc5bd68153e81
Stream
29748393-3905-47b0-878a-0c421ec3c428
Epoch
fbac9ea0-9c5d-48d1-8454-df1049493db7
Pool version
code_reasoning-v4
Checker version
code_reasoning-v2
Panel size
5
Provider spend
$0.0212818

Checker layer

Panel members.

# Item Item id Outcome Score
1 code-reasoning-015 0b253884-eae1-49ec-9926-0b5f7b4ea021 ok 1.00
2 code-reasoning-045 40c377f7-9280-4939-9cb3-4c3a858cef6b ok 1.00
3 code-reasoning-076 d490aa39-7087-4e12-9b8e-98f74d7e01c2 ok 1.00
4 code-reasoning-105 b5309579-c52d-4ad1-a049-017edbbc796a ok 1.00
5 code-reasoning-135 0471b2a9-c59a-4a89-a4ea-3fe357820c53 truncated 0.00

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 Item text stays unpublished so nobody can tune against the panel — identity travels as the item id and its pool source id.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-1787485214-OZpaNgbq2TjWBEDfyOBW stop 48088 188 1660 $0.0038074 981a451facac97ce5c9cec07596e9cdf1f1eb0c6bb43186a253e166800b41c07 2026-08-23 11:41:03 UTC
2 openrouter DeepInfra moonshotai/kimi-k2.5 gen-1787485214-Q0ynOQF2Lk5TeHlMtNwH stop 48888 286 1718 $0.0039942 0f7803889ef0ea8d7bbe5bed87fd9cc3af701a032bc36d5e8bfda48fe409df2e 2026-08-23 11:41:03 UTC
3 openrouter DeepInfra moonshotai/kimi-k2.5 gen-1787485214-JVvhDj4xyF5bMiSOhgT4 stop 53154 206 1907 $0.0043834 31b12bfd7061cb298b02289e1293852137ea14b712d2add9a3a8622c2c6bf755 2026-08-23 11:41:08 UTC
4 openrouter DeepInfra moonshotai/kimi-k2.5 gen-1787485214-cGVq4ascaazyq6iD1RuA stop 52472 215 1914 $0.0043911 b46f1b2467105c7b6d05d11994ef068d4f926a3363ee44f8fc1c2de9564f3641 2026-08-23 11:41:07 UTC
5 openrouter — — — — — 244 2048 $0.0047056 — 2026-08-23 11:42:21 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/36a4ce89e69d…

curl -s https://actuary.sh/api/v1/evidence/36a4ce89e69db47b493609110db3d7735fcebf3bbd5d7ca9b533619de96d8929 \
  | 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.