{"status":"live","updatedAt":"2026-08-06T12:57:29.158Z","benchmark":{"name":"CivicProofBench 100","version":"2.0.0","sha256":"8a8a9b1cdbd1596cc93cff186faa9d4181d8017a777a7462785a03f236cf71d6","cases":100},"counts":{"total":1,"pending":0,"verified":1,"rejected":0},"verifiedRuns":[{"kind":"external-model","rank":2,"runId":"github-models-mistral-ai-mistral-small-2503-30398743931","system":"mistral-ai/mistral-small-2503","provider":"GitHub Models","modelVersion":"1","evaluatedAt":"2026-07-28T21:00:31.351Z","benchmarkVersion":"2.0.0","exactReplay":34,"articleSetAccuracy":99,"dispositionAccuracy":55,"scoreMae":14.92,"evidenceUrl":"https://github.com/MARCCHERGGI/indie-metrics-site/tree/main/civicproof/results/mistral-small-2503-2026-07-28","workflowUrl":"https://github.com/MARCCHERGGI/indie-metrics-site/actions/runs/30398743931","independent":false,"verificationClass":"Project-maintainer CI run with complete public evidence; not independent third-party validation.","boundary":"Public open-book external-model result executed by CivicProof project-maintainer CI. It is reproducible evidence, not independent validation."}],"referenceRun":{"kind":"reference","rank":1,"runId":"CIVIC-PROOF-REFERENCE-2.0.0","system":"CIVIC-PROOF-1.0 reference engine","provider":"CivicProof","modelVersion":"deterministic lexical baseline","evaluatedAt":"2026-07-28T00:00:00.000Z","benchmarkVersion":"2.0.0","exactReplay":100,"articleSetAccuracy":100,"dispositionAccuracy":100,"scoreMae":0,"evidenceUrl":"/nation/leaderboard/reference","boundary":"Consistency baseline only. It reproduces its own public rules and is not an independent model result."},"boundary":"The Mistral row is a complete project-maintainer CI run, not independent third-party validation. Other external rows require human-verified public evidence. Pending identities and results remain private."}