{"created_at":"2026-10-05T09:12:57.912294Z","failure":null,"partial":false,"provenance":{"endpoint":{"base_url":"REDACTED_LOOPBACK_ENDPOINT","model":"glm53-csf-qad-tp2-dcp2-c4-327k-r1"},"finished_at":"2026-10-05T09:12:57.911894Z","started_at":"2026-10-05T09:12:34.145141Z","submitted_at":"2026-10-05T09:03:39.906249+00:00","worker":{"id":"REDACTED_RUN_OR_WORKER_ID"}},"results":{"evidence":{"completeness":"completed","created_at":"2026-10-05T09:12:57.900039Z","evidence_kind":"measured","failure":null,"identities":{"concurrency":{"requests":1,"workers":1},"context":{"configured_tokens":327680},"dataset":{"name":"deterministic-native","profile_sha256":"fce0e6e1e7497585bdfc6f76507d3baaff99a6b4f174d841d268de20c68de7f6"},"hardware":{"gpu":"2x RTX PRO 6000 Blackwell Max-Q","vram_mib_per_gpu":97887},"harnesses":{"native":{"schema":"anvil-serving.agentic/v1"}},"image":{"status":"not-applicable"},"model":{"repository":"local-inference-lab/GLM-5.3-Flash-NVFP4-MXFP8-CSF-QAD","revision":"4f90b74165cb1542cca35730b7dfa4d550bcf0ce","served_name":"glm53-csf-qad-tp2-dcp2-c4-327k-r1"},"runtime":{"engine_revision":"00a33e23142097f99fef95b1ecab6da4ea271e2e","image_id":"sha256:6008f020c23065bef71390194f968af9f5ebbedf57807b1659f876632e4a1067"},"served_model":{"configured_context":327680,"id":"glm53-csf-qad-tp2-dcp2-c4-327k-r1"},"topology":{"kind":"co-resident-benchmark-client","worker_id":"REDACTED_RUN_OR_WORKER_ID"}},"promotion":{"authorized":false,"message":"Benchmark evidence does not authorize model promotion; promotion is a separate human decision."},"run":{"ownership_id":"REDACTED_RUN_OR_WORKER_ID","profile":"scout","run_id":"REDACTED_RUN_OR_WORKER_ID","spec_sha256":"6514ba13cda6fd19b1272add1b8c519d22579d8f8df079840a83f7392bd0b4ab","suite":"agentic"},"schema":"anvil-serving.benchmark-evidence/v1","stages":[{"evidence":[{"bytes":551,"path":"evidence/0-assets.json","sha256":"3d3c75f915e9e45a85827741975750fb389a44c092ff0da3121b87a49f64569f"}],"name":"asset_preparation","sequence":0,"status":"completed"},{"evidence":[{"bytes":2001,"path":"evidence/0-preflight.json","sha256":"a0dba29ffc1f7975dcdf263a007165624b858c6eb7bf6e2b4ac64b8e9462f856"}],"name":"preflight","sequence":1,"status":"completed"},{"evidence":[{"bytes":36742,"path":"evidence/0-agentic.json","sha256":"a081a35bce3661ac8aa0715d75addd3855b52bac7ca9297e2cbd75314f8c8c67"}],"name":"agentic","sequence":2,"status":"completed"}],"summary":{"attempted":18,"pass_rate":0.9444444444444444,"passed":17,"required_pass_rate":0.75}}},"run":{"ownership_id":"REDACTED_RUN_OR_WORKER_ID","profile":"scout","run_id":"REDACTED_RUN_OR_WORKER_ID","spec_sha256":"6514ba13cda6fd19b1272add1b8c519d22579d8f8df079840a83f7392bd0b4ab","suite":"agentic"},"schema":"anvil-serving.benchmark-result/v1","status":"completed"}
