Actuary
← Meta Llama 3.3 70B Instruct

Evidence block

Code reasoning · cycle 43

meta-llama/llama-3.3-70b-instruct · 2026-08-05 13:28:57 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
0.000
Members
5
Checker scored
5
Frozen floor
0.700
Anytime level α
0.050
Hashes and versions
Evidence hash
25353d276e4b4fe171b5def153182584eeb5eb2310fc50a4e170c574f014a552
Design hash
bc3668bd52d9d48803b744c2c90f4ed816d7e394525b32b4219f6bab5c9c6df2
Panel hash
827de544c7641d140ac37e321bdb2578a41e291915f519183abbc5bd68153e81
Stream
df8bada5-ce30-4228-8edb-a0ecaac008e4
Epoch
d023180f-6535-4cc4-8b73-c6060ce2b87a
Pool version
code_reasoning-v4
Checker version
code_reasoning-v2
Panel size
5
Provider spend
$0.0001405

Checker layer

Panel members.

# Item Item id Outcome Score
1 code-reasoning-015 0b253884-eae1-49ec-9926-0b5f7b4ea021 ok 0.00
2 code-reasoning-045 40c377f7-9280-4939-9cb3-4c3a858cef6b ok 0.00
3 code-reasoning-076 d490aa39-7087-4e12-9b8e-98f74d7e01c2 ok 0.00
4 code-reasoning-105 b5309579-c52d-4ad1-a049-017edbbc796a ok 0.00
5 code-reasoning-135 0471b2a9-c59a-4a89-a4ea-3fe357820c53 ok 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 Item text stays unpublished so nobody can tune against the panel — identity travels as the item id and its pool source id.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 meta-llama/llama-3.3-70b-instruct gen-1785936531-W2kxFsQKTyr0iLmPMCrr stop 714 187 7 $0.0000209 7510fc43e245ba983598cf583dfe0fa072008b740982a14508e9ab3eccb4d0a4 2026-08-05 13:28:52 UTC
2 openrouter DeepInfra meta-llama/llama-3.3-70b-instruct gen-1785936530-e8FjBoo2Xbz9t2drFhrb stop 1225 287 16 $0.0000338 0eb7772b438a5c4b83f858b4271aac90dec41823eb6215950ab80a9573bf3721 2026-08-05 13:28:51 UTC
3 openrouter DeepInfra meta-llama/llama-3.3-70b-instruct gen-1785936531-aK1dN9QBJrRRieE73ATY stop 3184 207 21 $0.0000274 053e68afc0dc61cd93f64dd3bb2b789b528de55447e01564888df1219d2d11aa 2026-08-05 13:28:53 UTC
4 openrouter DeepInfra meta-llama/llama-3.3-70b-instruct gen-1785936530-75DayfMgxoKuMQQrIm3t stop 2740 214 14 $0.0000259 bbc50681639577333111b3896d62412f74f65cfc17bc0ae3f4e673816c28188c 2026-08-05 13:28:53 UTC
5 openrouter DeepInfra meta-llama/llama-3.3-70b-instruct gen-1785936535-vGmxTGZb5hxCyaineXuf stop 1506 244 25 $0.0000324 a94e2966d81db95231dd9cf8abeeaeab69b9338b27c9edd784f662e44fb35680 2026-08-05 13:28:57 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/25353d276e4b…

curl -s https://actuary.sh/api/v1/evidence/25353d276e4b4fe171b5def153182584eeb5eb2310fc50a4e170c574f014a552 \
  | 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.