{"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/bca84a08-d1f5-5ff9-83db-659a561aba75","source_aggregate":"/data/releases/2026.08-frontier-knowledge-v4/fmb-live-runs-capstone.json"}},"run":{"benchmarkId":"FMB-CAP-ACC-03","benchmarkVersion":"0.3.0-dev","campaignId":"367b3b25-259f-5466-8e2f-1b8faa2b4365","comparisonGroupId":"accounting-capstone-0.3.0-dev-configuration-specific-v1","comparisonProtocol":"configuration_specific","completedAt":"2026-07-27T04:39:10.372733Z","configuration":{"comparisonProtocol":"configuration_specific","key":"config-58353b1a3da1d308bba6fcd143d0808790725b421eaac69aaf3489a81840a2cf","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-v8","toolAccess":"none"},"displayVersion":"v0.2","execution":{"attempted":100,"invalidJson":0,"missing":0,"providerFailed":0,"schemaFailed":0,"tasksTotal":100,"valid":100},"globalReleaseId":"c44af57d-9091-45c8-9299-96dee423e175","judge":{"assessmentNotRequiredTasks":0,"assessmentTasks":100,"costUsd":"3.080510125","modelId":"openai/gpt-5.6-sol","providerName":"OpenRouter (Azure EU)","providerRegion":null,"retries":0,"usage":{"inputTokens":242177,"outputTokens":52868,"totalTokens":295045}},"manifest":{"generatedAt":"2026-08-14T12:57:07Z","sha256":"caa004fcccade516afe661a318e69b555844595cbcd726cdc749ff3c64a6378f"},"modelDisplayName":"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":"bca84a08-d1f5-5ff9-83db-659a561aba75","score":{"earned":"742.42380952381","maximum":"1000","percentage":"74.2424"},"scorerId":"fmb-live-scorer-v1","scoringMethod":"hybrid","selectionState":"primary","signedReleaseId":"9d07d6d8-80f4-4bca-9adf-1f436336d09d","sourceAggregate":{"campaignId":"367b3b25-259f-5466-8e2f-1b8faa2b4365","path":"fmb-live-runs-capstone.json","sha256":"85402a8acfbe9f2ff893e95de5c1e105daada9d13ebf267a87fc67d858020660","sizeBytes":21048},"startedAt":"2026-07-26T19:17:45.555338Z","suiteId":"accounting-capstone","suiteTitle":"Accounting Capstone","testedModelCostUsd":"0.48488","toolAccess":"none","totalCostUsd":"3.565390125","usage":{"inputTokens":130750,"outputTokens":42670,"totalTokens":173420},"validity":"provisional"},"schema_version":"kaelum-bench-public-run-projection-v1"}
