{"provenance":{"authentication":"derived_from_authenticated_public_release","separately_signed":false,"statement":"This sanitized run projection is derived from the authenticated public release and is not separately signed.","urls":{"manifest":"/data/releases/2026.08-frontier-knowledge-v4/public-manifest.json","manifest_signature":"/data/releases/2026.08-frontier-knowledge-v4/public-manifest-signature.json","public_keys":[{"key_id":"kaelum-bench-public-release-20260718-v1","url":"/data/keys/kaelum-bench-public-release-20260718-v1"}],"run_index":"/data/releases/2026.08-frontier-knowledge-v4/public-run-index.json","self":"/data/runs/30e5711c-94ae-56f8-bd66-a5fa484b2fd4","source_aggregate":"/data/releases/2026.08-frontier-knowledge-v4/fmb-live-runs.json"}},"run":{"benchmarkId":"FMB-KNOW-ACC-01","benchmarkVersion":"0.1.0-dev","campaignId":"c5f9dc7e-c31c-44ef-bbce-e6eb1c4775d0","comparisonGroupId":"accounting-knowledge-0.1.0-dev-configuration-specific-v1","comparisonProtocol":"configuration_specific","completedAt":"2026-07-16T06:40:51.135751Z","configuration":{"comparisonProtocol":"configuration_specific","key":"config-a5640a517f74f5215c60166d203ddb0defd169cc6908c9fb8c3a495ec7182784","materialParameters":[{"name":"allow_fallbacks","value":"false"},{"name":"data_collection","value":"deny"},{"name":"max_attempts_per_task","value":"1"},{"name":"max_concurrency","value":"1"},{"name":"max_generation_metadata_lookups","value":"1"},{"name":"max_output_tokens","value":"34000"},{"name":"max_raw_response_bytes","value":"16777216"},{"name":"max_request_bytes","value":"1048576"},{"name":"metadata_timeout_seconds","value":"60"},{"name":"request_timeout_seconds","value":"300"},{"name":"require_parameters","value":"true"},{"name":"temperature","value":"0"},{"name":"zdr","value":"true"}],"providerRoute":"openrouter/xai","reasoningEffort":"high","reasoningProfile":"reasoning-max-supported","requestedModelId":"x-ai/grok-4.5","runtimeMode":"openrouter-tool-free-non-streaming-v4","toolAccess":"none"},"displayVersion":"v0.2","execution":{"attempted":60,"invalidJson":0,"missing":0,"providerFailed":0,"schemaFailed":0,"tasksTotal":60,"valid":60},"globalReleaseId":"c44af57d-9091-45c8-9299-96dee423e175","judge":{"assessmentNotRequiredTasks":0,"assessmentTasks":60,"costUsd":"0.86528","modelId":"openai/gpt-5.6-sol","providerName":"OpenRouter (Azure)","providerRegion":null,"retries":58,"usage":{"inputTokens":45490,"outputTokens":21261,"totalTokens":66751}},"manifest":{"generatedAt":"2026-08-14T12:57:07Z","sha256":"caa004fcccade516afe661a318e69b555844595cbcd726cdc749ff3c64a6378f"},"modelDisplayName":"xAI: Grok 4.5","providerLabel":"xAI","reasoningEffort":"high","releaseSignature":{"algorithm":"ed25519","keyId":"kaelum-bench-public-release-20260718-v1","messageType":"kaelum.bench.fmb.public_release.v1","payloadSha256":"caa004fcccade516afe661a318e69b555844595cbcd726cdc749ff3c64a6378f","signedFile":"public-manifest.json"},"requestedModelId":"x-ai/grok-4.5","revisionState":"original","runId":"30e5711c-94ae-56f8-bd66-a5fa484b2fd4","score":{"earned":"575.523809523809","maximum":"600","percentage":"95.9206"},"scorerId":"fmb-live-scorer-v1","scoringMethod":"hybrid","selectionState":"primary","signedReleaseId":"0bae13a5-a0c9-44fb-b6df-6bc83a6847d5","sourceAggregate":{"campaignId":"c5f9dc7e-c31c-44ef-bbce-e6eb1c4775d0","path":"fmb-live-runs.json","sha256":"2bd457a3aa6457408e920ff34edb9733d3296fd252ebc4d80e8ee1986db72119","sizeBytes":18268},"startedAt":"2026-07-16T06:15:23.081353Z","suiteId":"accounting-knowledge","suiteTitle":"Accounting Knowledge","testedModelCostUsd":"0.216928","toolAccess":"none","totalCostUsd":"1.082208","usage":{"inputTokens":60110,"outputTokens":19446,"totalTokens":79556},"validity":"provisional"},"schema_version":"kaelum-bench-public-run-projection-v1"}
