{"batch_id":"2026-07-22-pilot-b-deepseek-v4-flash-a","schema_version":"1.0","study":"pilot-b","traces":[{"architecture":"contracted_react","budget":{"max_concurrent_branches":3,"max_model_calls":18,"max_tool_calls":24,"max_total_tokens":30000,"timeout_seconds":240},"eligibility_class":"eligible_agent_result","expected_model_id":"deepseek-v4-flash","family":"alternative_path","metrics":{"completion_tokens":137,"controller_rejection_count":0,"critical_path_rounds":4,"elapsed_seconds":5.018038899987005,"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":0,"model_calls":5,"path_switch":true,"prompt_tokens":2659,"repeated_action_count":0,"required_fact_recall":1.0,"tool_calls":5,"total_tokens":2796,"unrelated_action_count":0,"utility":100},"model_exchanges":[{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.032197599997744,"model_round":1,"prompt_sha256":"72acd4fb9649938f93bb1039b5fab440ea9d816649bdfefd7ef37b0a7f544477","provider_model":"deepseek-v4-flash","response_sha256":"1199fd234b315946d0650fb28abc53628005ff2fa289c4986b1b55c67fc6af9b","usage":{"completion":16,"prompt":388,"total":404}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.1030932000139728,"model_round":2,"prompt_sha256":"6e5eda99c15c3943a5c6d4ea3d4580d2d0b1250a31c3d0514cbc5dd593e362f8","provider_model":"deepseek-v4-flash","response_sha256":"295fdef966214727a1013e96c0eb72f6bbcc1f5d8d5dfb5e6face00863de3983","usage":{"completion":28,"prompt":461,"total":489}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.9525211999425665,"model_round":3,"prompt_sha256":"2f5824f5fb54f30a060a37bec2ec9f902952a30a3a767452a22851444e780fdf","provider_model":"deepseek-v4-flash","response_sha256":"5ba84ad90f55e4808ccf8e4c18893f24557b94e3851892cf5f07b67f85eb50fe","usage":{"completion":29,"prompt":515,"total":544}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.1061766999773681,"model_round":4,"prompt_sha256":"d44a2169948be779c105d862fe2fdce2b565e2594d0069f9d2a5011f09b81dfa","provider_model":"deepseek-v4-flash","response_sha256":"203d8619b22a85be1124e4a141d0f478c94c94d64b6b9394e7a0c03b6a3c2a92","usage":{"completion":49,"prompt":608,"total":657}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.8225448000011966,"model_round":5,"prompt_sha256":"626bf4aab1bb57f09f70d0706cd7dcaf7d445f66133779b5739a09427acfad97","provider_model":"deepseek-v4-flash","response_sha256":"7310372e895dfcd49057382944f726a4b0ab03cbdea6904b834b3516d3bead52","usage":{"completion":15,"prompt":687,"total":702}}],"observed_model_ids":["deepseek-v4-flash"],"pair_id":"alternative_path-01","requested_model_id":"deepseek-v4-flash","run_key":"contracted_react__alternative_path-01","trace":[{"action":{"tool":"list_evidence"},"branch_id":"controller","disposition":"executed","model_round":1,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":1},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"tool_error","model_round":2,"observation":{"error_code":"route_unavailable","fact_count":0,"ok":false,"state_changed":true},"step":2},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":3,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":3},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"executed","model_round":4,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":4},{"action":{"tool":"finish"},"branch_id":"controller","disposition":"executed","model_round":5,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":5}],"verification":{"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":0,"required_fact_recall":1.0,"utility":100,"verified_success":true}},{"architecture":"goal_graph_hybrid","budget":{"max_concurrent_branches":3,"max_model_calls":18,"max_tool_calls":24,"max_total_tokens":30000,"timeout_seconds":240},"eligibility_class":"eligible_agent_result","expected_model_id":"deepseek-v4-flash","family":"alternative_path","metrics":{"completion_tokens":280,"controller_rejection_count":3,"critical_path_rounds":4,"elapsed_seconds":8.094181300024502,"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":0,"model_calls":8,"path_switch":true,"prompt_tokens":4375,"repeated_action_count":3,"required_fact_recall":1.0,"tool_calls":5,"total_tokens":4655,"unrelated_action_count":0,"utility":100},"model_exchanges":[{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.886095599969849,"model_round":1,"prompt_sha256":"04ff39e755b40b22cc16c72b931a6235f5502c9b8c30f0a9f957f1fe0b9f4570","provider_model":"deepseek-v4-flash","response_sha256":"1199fd234b315946d0650fb28abc53628005ff2fa289c4986b1b55c67fc6af9b","usage":{"completion":16,"prompt":519,"total":535}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.065752699971199,"model_round":2,"prompt_sha256":"ea9c33161ec96aacb0634bf79ddd27192366af53eeeefc82dfc967f1229bfce9","provider_model":"deepseek-v4-flash","response_sha256":"295fdef966214727a1013e96c0eb72f6bbcc1f5d8d5dfb5e6face00863de3983","usage":{"completion":28,"prompt":508,"total":536}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.7909503999399021,"model_round":3,"prompt_sha256":"94034cf43a7ca8a184ae11cf581675f20d6cd3b738dd120f3ee790b2eb133127","provider_model":"deepseek-v4-flash","response_sha256":"5ba84ad90f55e4808ccf8e4c18893f24557b94e3851892cf5f07b67f85eb50fe","usage":{"completion":29,"prompt":515,"total":544}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.9399859000695869,"model_round":4,"prompt_sha256":"6a3628b53697f3b93ef65920ad9fe2a70b7a6e0d5ee2828239f0efb499463a9b","provider_model":"deepseek-v4-flash","response_sha256":"203d8619b22a85be1124e4a141d0f478c94c94d64b6b9394e7a0c03b6a3c2a92","usage":{"completion":49,"prompt":549,"total":598}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.223781400010921,"model_round":5,"prompt_sha256":"19d242e07a398be1313d1392d30f86035bb8e730d3cd6d81c99391b1b02bf26f","provider_model":"deepseek-v4-flash","response_sha256":"203d8619b22a85be1124e4a141d0f478c94c94d64b6b9394e7a0c03b6a3c2a92","usage":{"completion":49,"prompt":538,"total":587}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.152879000059329,"model_round":6,"prompt_sha256":"46522d73c8751dc91c25db3d3d298efe36b3b137a8494f3f087cf6be52307e86","provider_model":"deepseek-v4-flash","response_sha256":"203d8619b22a85be1124e4a141d0f478c94c94d64b6b9394e7a0c03b6a3c2a92","usage":{"completion":49,"prompt":560,"total":609}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.1385329000186175,"model_round":7,"prompt_sha256":"304ad45af8ded4b09f25a056cba7fc270002ebd50701ec5470c7f237d531e304","provider_model":"deepseek-v4-flash","response_sha256":"843caf795919aa6a0c0052ce9f76769f21efcf847577f788f994d3f6ee5db258","usage":{"completion":45,"prompt":582,"total":627}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.8932966999709606,"model_round":8,"prompt_sha256":"36debcc9dccbd6ae71b2fdb1b1ccfdcba6ce9f137a888462efbd12b8667bbd2a","provider_model":"deepseek-v4-flash","response_sha256":"7310372e895dfcd49057382944f726a4b0ab03cbdea6904b834b3516d3bead52","usage":{"completion":15,"prompt":604,"total":619}}],"observed_model_ids":["deepseek-v4-flash"],"pair_id":"alternative_path-01","requested_model_id":"deepseek-v4-flash","run_key":"goal_graph_hybrid__alternative_path-01","trace":[{"action":{"tool":"list_evidence"},"branch_id":"controller","disposition":"executed","model_round":1,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":1},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"tool_error","model_round":2,"observation":{"error_code":"route_unavailable","fact_count":0,"ok":false,"state_changed":true},"step":2},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":3,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":3},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"executed","model_round":4,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":4},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":5,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":5},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":6,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":6},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":7,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":7},{"action":{"tool":"finish"},"branch_id":"controller","disposition":"executed","model_round":8,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":8}],"verification":{"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":0,"required_fact_recall":1.0,"utility":100,"verified_success":true}},{"architecture":"contracted_react","budget":{"max_concurrent_branches":3,"max_model_calls":18,"max_tool_calls":24,"max_total_tokens":30000,"timeout_seconds":240},"eligibility_class":"eligible_agent_result","expected_model_id":"deepseek-v4-flash","family":"open_evidence","metrics":{"completion_tokens":198,"controller_rejection_count":0,"critical_path_rounds":5,"elapsed_seconds":5.13136769994162,"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":0,"model_calls":6,"path_switch":false,"prompt_tokens":4120,"repeated_action_count":0,"required_fact_recall":1.0,"tool_calls":6,"total_tokens":4318,"unrelated_action_count":0,"utility":100},"model_exchanges":[{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.7625149000668898,"model_round":1,"prompt_sha256":"091caacb8265f87c1ad6255d34d58c205183e143fd550de3bc9a56f3a0219bc7","provider_model":"deepseek-v4-flash","response_sha256":"1199fd234b315946d0650fb28abc53628005ff2fa289c4986b1b55c67fc6af9b","usage":{"completion":16,"prompt":435,"total":451}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.8152038999833167,"model_round":2,"prompt_sha256":"46a0e0876175c6456f3d5aa22ac0e2e6fabe752cc63c72b0cabcc6879f967cce","provider_model":"deepseek-v4-flash","response_sha256":"83974279ee363b0f44ba453803a6840d09132b3dfa433f960fb6edb1fb9d1136","usage":{"completion":28,"prompt":553,"total":581}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.0001199000980705,"model_round":3,"prompt_sha256":"e347de150e6b1433bc12a3471321d0c595250cc17cfe3fff947d782b7340441d","provider_model":"deepseek-v4-flash","response_sha256":"f78bf6fcf38377ff9cf4bcaf2b8bedc28bec540921b67a059fc3ba69b4c795c8","usage":{"completion":29,"prompt":642,"total":671}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.9666550999972969,"model_round":4,"prompt_sha256":"97bdc1f020654e706790eb3d7a2b5c0be98eba723f621051f98d8f4f045d902d","provider_model":"deepseek-v4-flash","response_sha256":"766478af48faa1ae324506451d32fdb13724e9f2a6864bbc04c0b06c847044e5","usage":{"completion":29,"prompt":735,"total":764}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.952361400006339,"model_round":5,"prompt_sha256":"7df8cb58a8b57737018663dc78c1111d3f66bb544ebfb97f6b20a3e5a71b967e","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":822,"total":903}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.6326592999976128,"model_round":6,"prompt_sha256":"027a9948bc29cfaf805f2b4e43816003549e489124bb235f61a2ecd4a803de37","provider_model":"deepseek-v4-flash","response_sha256":"7310372e895dfcd49057382944f726a4b0ab03cbdea6904b834b3516d3bead52","usage":{"completion":15,"prompt":933,"total":948}}],"observed_model_ids":["deepseek-v4-flash"],"pair_id":"open_evidence-01","requested_model_id":"deepseek-v4-flash","run_key":"contracted_react__open_evidence-01","trace":[{"action":{"tool":"list_evidence"},"branch_id":"controller","disposition":"executed","model_round":1,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":1},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":2,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":2},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":3,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":3},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":4,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":4},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"executed","model_round":5,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":5},{"action":{"tool":"finish"},"branch_id":"controller","disposition":"executed","model_round":6,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":6}],"verification":{"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":0,"required_fact_recall":1.0,"utility":100,"verified_success":true}},{"architecture":"goal_graph_hybrid","budget":{"max_concurrent_branches":3,"max_model_calls":18,"max_tool_calls":24,"max_total_tokens":30000,"timeout_seconds":240},"eligibility_class":"eligible_agent_result","expected_model_id":"deepseek-v4-flash","family":"open_evidence","metrics":{"completion_tokens":1231,"controller_rejection_count":13,"critical_path_rounds":5,"elapsed_seconds":18.76757619995624,"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":0,"model_calls":18,"path_switch":false,"prompt_tokens":13989,"repeated_action_count":13,"required_fact_recall":1.0,"tool_calls":5,"total_tokens":15220,"unrelated_action_count":0,"utility":100},"model_exchanges":[{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.6368815000168979,"model_round":1,"prompt_sha256":"ad0fc9e5a3ebd3522e0e8e3bd4b4ed5d14f9252c9c382650bab6ba698809ca7b","provider_model":"deepseek-v4-flash","response_sha256":"0929f8a7ea7640da08e12673da0f1f33c8572bd3cf1f7721e9a0e0984916ada1","usage":{"completion":11,"prompt":657,"total":668}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.8402027999982238,"model_round":2,"prompt_sha256":"8060f21995bbda73bfd4f889ba7bef9ecb1a3f300beec5ebcd05408ec5646cca","provider_model":"deepseek-v4-flash","response_sha256":"83974279ee363b0f44ba453803a6840d09132b3dfa433f960fb6edb1fb9d1136","usage":{"completion":28,"prompt":646,"total":674}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.832384999957867,"model_round":3,"prompt_sha256":"2fd7e16a27a83c2ea6e7a083c72cc32a476fb1f446f721f90348adccfd86fe99","provider_model":"deepseek-v4-flash","response_sha256":"f78bf6fcf38377ff9cf4bcaf2b8bedc28bec540921b67a059fc3ba69b4c795c8","usage":{"completion":29,"prompt":657,"total":686}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.8088202000362799,"model_round":4,"prompt_sha256":"25cb5e66517377947084381d98e5d32f4f441016ed37a112e3bd2811c7405a23","provider_model":"deepseek-v4-flash","response_sha256":"766478af48faa1ae324506451d32fdb13724e9f2a6864bbc04c0b06c847044e5","usage":{"completion":29,"prompt":670,"total":699}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.8900903000030667,"model_round":5,"prompt_sha256":"e18e4c3a0116cf57cc3d8adafcc0eead94f4e4c71b8f2f10b7752f51fb386433","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":699,"total":780}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.0002814000472426,"model_round":6,"prompt_sha256":"8d57bd8b3def283739d1557d88db847172da131dc97841243c8d008d28721769","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":688,"total":769}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.828512700041756,"model_round":7,"prompt_sha256":"1dc7e8a22bd891dd67f8b1042c50a7ad02f02e2d713a4d724b9871d660481d4c","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":710,"total":791}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.187772999983281,"model_round":8,"prompt_sha256":"bc972f4b612b5e013d4171f728c8e79cddbfdcb9c2b292f3a4c86c7e2310e31b","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":732,"total":813}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.3038637000136077,"model_round":9,"prompt_sha256":"550171a5e4a779a098586d2175d31635e392156f955ae31ba16fd34c1f0ee0b4","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":754,"total":835}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.2092899000272155,"model_round":10,"prompt_sha256":"ffe17e307004747b750d42005809feb2b0c5383787f55846096c8d02f978cf5a","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":776,"total":857}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.2458027000539005,"model_round":11,"prompt_sha256":"3f3a9409271867981c6a5900fc7c5bdcf694b0ae5a2151b32c89102d5f8fd5e5","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":798,"total":879}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.3113930999534205,"model_round":12,"prompt_sha256":"3a9943dcc3ce4176519f6c501c8b7275f0ad09b2fc08520c9e3286f97aaf7c03","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":820,"total":901}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.2096201999811456,"model_round":13,"prompt_sha256":"839bf2e98d3182bcd91ebd9d87dbc7cf16889bf66aced480ce9dd6d46c5610e2","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":842,"total":923}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.9447885999688879,"model_round":14,"prompt_sha256":"f77b32c25754d21fef9793b7c4d9f7ca6de867756c5f459e82b6ee42c524057f","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":864,"total":945}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.177284499979578,"model_round":15,"prompt_sha256":"e4e9d6e710e073f9235424b07c9158c3be6bd6bd185ef1c0ad5a653fdef01a93","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":886,"total":967}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.216277200030163,"model_round":16,"prompt_sha256":"20f7b6be9ac7cef3ee9c9c837f83a4077b1cb094e9307d5f35f26832bc448198","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":908,"total":989}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.1039398000575602,"model_round":17,"prompt_sha256":"e33f1f9ad0dafcd7249e0a8b1ea187fe08dcc989a9c16fc44b05a34c3a20f976","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":930,"total":1011}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.0143326000543311,"model_round":18,"prompt_sha256":"29fdced443a60758a23e8427ca23f2e1b9f5dad9eb2725e78b528bc6577dc374","provider_model":"deepseek-v4-flash","response_sha256":"16a25e35fee767802b2919d79136ea77f5c2a258684d02865f452c6c1524b4c0","usage":{"completion":81,"prompt":952,"total":1033}}],"observed_model_ids":["deepseek-v4-flash"],"pair_id":"open_evidence-01","requested_model_id":"deepseek-v4-flash","run_key":"goal_graph_hybrid__open_evidence-01","trace":[{"action":{"tool":"list_evidence"},"branch_id":"controller","disposition":"executed","model_round":1,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":1},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":2,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":2},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":3,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":3},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":4,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":4},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"executed","model_round":5,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":5},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":6,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":6},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":7,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":7},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":8,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":8},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":9,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":9},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":10,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":10},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":11,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":11},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":12,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":12},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":13,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":13},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":14,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":14},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":15,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":15},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":16,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":16},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":17,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":17},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":18,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":18}],"verification":{"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":0,"required_fact_recall":1.0,"utility":100,"verified_success":true}},{"architecture":"contracted_react","budget":{"max_concurrent_branches":3,"max_model_calls":18,"max_tool_calls":24,"max_total_tokens":30000,"timeout_seconds":240},"eligibility_class":"eligible_agent_result","expected_model_id":"deepseek-v4-flash","family":"parallel_decomposition","metrics":{"completion_tokens":206,"controller_rejection_count":0,"critical_path_rounds":5,"elapsed_seconds":6.456444300012663,"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":3,"model_calls":6,"path_switch":false,"prompt_tokens":3813,"repeated_action_count":0,"required_fact_recall":1.0,"tool_calls":6,"total_tokens":4019,"unrelated_action_count":0,"utility":25},"model_exchanges":[{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.0167829000856727,"model_round":1,"prompt_sha256":"f0bab67495651ae4bcb22f53257d8af4d777f8d78e07a18a48010d99e16bc204","provider_model":"deepseek-v4-flash","response_sha256":"1199fd234b315946d0650fb28abc53628005ff2fa289c4986b1b55c67fc6af9b","usage":{"completion":16,"prompt":416,"total":432}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.8965952999424189,"model_round":2,"prompt_sha256":"d6e346682f3cb8b594d6de97f135f965a9c17d920cfc4f1425dd385e50331eaa","provider_model":"deepseek-v4-flash","response_sha256":"1de8849ca69262c0e648cf36bd05d8d9581a3545f073721fad91bb36954de3ca","usage":{"completion":28,"prompt":503,"total":531}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.0361864999867976,"model_round":3,"prompt_sha256":"83fae745b1f9d5dffefdec588a739ec73e5dfb75c64898b184e09ce8a4e55e0a","provider_model":"deepseek-v4-flash","response_sha256":"f13f77122b5d1af690bab4fe8c29a53762f4a826f918eabb67645da48c459cbd","usage":{"completion":28,"prompt":587,"total":615}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.9923233999870718,"model_round":4,"prompt_sha256":"e8d9496b3b3b020905f9680d7d1a28aef1579139c914bca569b6e9dc6891a0d0","provider_model":"deepseek-v4-flash","response_sha256":"f66599ca184ac1930b0d5761d8c44ad03628cb1aef77d215737de5d774ffcdcb","usage":{"completion":29,"prompt":671,"total":700}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.4305322000291198,"model_round":5,"prompt_sha256":"1d59c49d5d81743e26a6f19907b873a4ca23e11f3cb8ba152b0c3dc0ccc25862","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":758,"total":848}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.0821566000813618,"model_round":6,"prompt_sha256":"eb5d48c85da85b0c5d7ea1bbb53eb4db1ede917c393d96a68188106447bc773c","provider_model":"deepseek-v4-flash","response_sha256":"7310372e895dfcd49057382944f726a4b0ab03cbdea6904b834b3516d3bead52","usage":{"completion":15,"prompt":878,"total":893}}],"observed_model_ids":["deepseek-v4-flash"],"pair_id":"parallel_decomposition-01","requested_model_id":"deepseek-v4-flash","run_key":"contracted_react__parallel_decomposition-01","trace":[{"action":{"tool":"list_evidence"},"branch_id":"controller","disposition":"executed","model_round":1,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":1},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":2,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":2},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":3,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":3},{"action":{"tool":"read_evidence"},"branch_id":"controller","disposition":"executed","model_round":4,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":4},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"executed","model_round":5,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":5},{"action":{"tool":"finish"},"branch_id":"controller","disposition":"executed","model_round":6,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":6}],"verification":{"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":3,"required_fact_recall":1.0,"utility":25,"verified_success":false}},{"architecture":"goal_graph_hybrid","budget":{"max_concurrent_branches":3,"max_model_calls":18,"max_tool_calls":24,"max_total_tokens":30000,"timeout_seconds":240},"eligibility_class":"eligible_agent_result","expected_model_id":"deepseek-v4-flash","family":"parallel_decomposition","metrics":{"completion_tokens":1328,"controller_rejection_count":13,"critical_path_rounds":2,"elapsed_seconds":20.550333999912255,"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":3,"model_calls":18,"path_switch":false,"prompt_tokens":13426,"repeated_action_count":13,"required_fact_recall":1.0,"tool_calls":5,"total_tokens":14754,"unrelated_action_count":0,"utility":25},"model_exchanges":[{"branch_id":"alpha","error_code":null,"finish_reason":"stop","latency_seconds":1.0336132999509573,"model_round":1,"prompt_sha256":"abf4be44b6597f6e4306afff3a30c63892ea7717a4a6331fb9431c04c654c245","provider_model":"deepseek-v4-flash","response_sha256":"9e7968c4d43b8acfa2488237b84398061841be61d4b7ac91dc30f0047173c0bb","usage":{"completion":17,"prompt":678,"total":695}},{"branch_id":"beta","error_code":null,"finish_reason":"stop","latency_seconds":0.9532364000333473,"model_round":1,"prompt_sha256":"f2718f4e414bf87491b96e6de655150318e1bacae142230b29f5ffd2fbfb224d","provider_model":"deepseek-v4-flash","response_sha256":"14f1f147551bfe49d431a2dfc29768d532d90b992fb959c08cb324b269a6ccb9","usage":{"completion":17,"prompt":678,"total":695}},{"branch_id":"gamma","error_code":null,"finish_reason":"stop","latency_seconds":0.9596955000888556,"model_round":1,"prompt_sha256":"b7ac62f2b6d734cbf2518069850031ed88233af3eda64d73254836ad42800b87","provider_model":"deepseek-v4-flash","response_sha256":"c65f6c95f348f2420eff8c516a08897529fadaec1456a4ca1256f6b1ce59245e","usage":{"completion":18,"prompt":679,"total":697}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.2537893999833614,"model_round":2,"prompt_sha256":"afa95fa19f53563fc186b764c8b0d1e7cefb6b9e0074452841f4f70828c37365","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":656,"total":746}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.360450500040315,"model_round":3,"prompt_sha256":"54bc6c660a03e54b705304fc8fbe83dca2b8179fed0e5435d010826e35abf15f","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":645,"total":735}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.398023199988529,"model_round":4,"prompt_sha256":"86bf64e6213639976553f7807656911d176f90b0a1ae523454574d69b777ba1a","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":667,"total":757}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.419282600050792,"model_round":5,"prompt_sha256":"49a733b2c2ca4ba30bd2beb08191308477637d31c0221836c66c4c665c631243","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":689,"total":779}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.2220721999183297,"model_round":6,"prompt_sha256":"c9fd3a36f4a4b5ca1ab0c27b826e5857c827bd920c4086a3c45eafe2ea4c6c37","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":711,"total":801}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":0.9286682000383735,"model_round":7,"prompt_sha256":"73140a882a4f8ee3dd1aec9e827cd6b409b02d55779a3b39a70dbf6fbb5ab5ad","provider_model":"deepseek-v4-flash","response_sha256":"1199fd234b315946d0650fb28abc53628005ff2fa289c4986b1b55c67fc6af9b","usage":{"completion":16,"prompt":733,"total":749}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.3030839000130072,"model_round":8,"prompt_sha256":"0938e708ba0a84035d4de6c77bd0bc896ff56747f44fb181e49c19b908e05f05","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":722,"total":812}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.3230098999338225,"model_round":9,"prompt_sha256":"8d49a2d6a8a3c6ca049cffed328742f6ec5ee6c606f8a0f0ca741dfff23a94cf","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":744,"total":834}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.2809482000302523,"model_round":10,"prompt_sha256":"cf00744a98027af39ba363fe0ce21fa06b005a9111b10602f1e99f809e985d73","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":766,"total":856}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.338355700019747,"model_round":11,"prompt_sha256":"fe99616c480af6fee63b85e48f723df37be499d00910bbf61e05893d9c23f8fc","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":788,"total":878}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.2463418000843376,"model_round":12,"prompt_sha256":"ccf711ba717f60411d21538ba4e4be2be602f0d2ea74eff8a3d5780b84cc7d33","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":810,"total":900}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.344564399914816,"model_round":13,"prompt_sha256":"a67d601be45fcb166853d14b55767cc8847b5572233425c1d71ceda57a247663","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":832,"total":922}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.4053690999280661,"model_round":14,"prompt_sha256":"0683bf2181cbc621357724815b6f200fbb1468ffc9babde39a4a367c4331d24a","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":854,"total":944}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.3435086000245064,"model_round":15,"prompt_sha256":"8b24543e175ff8e7bd610ed3ee2265f71ebd1012ecdf096dc50df156f24c9cbe","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":876,"total":966}},{"branch_id":"controller","error_code":null,"finish_reason":"stop","latency_seconds":1.342006399994716,"model_round":16,"prompt_sha256":"5829d8da12b9832f96d5338c5704a293e3508491beaf894cf3ac6c534232f005","provider_model":"deepseek-v4-flash","response_sha256":"0e432736d38ed622dfda6f4bd17ee4a286d5a3d19cd34db97a72ba2c45ef0910","usage":{"completion":90,"prompt":898,"total":988}}],"observed_model_ids":["deepseek-v4-flash"],"pair_id":"parallel_decomposition-01","requested_model_id":"deepseek-v4-flash","run_key":"goal_graph_hybrid__parallel_decomposition-01","trace":[{"action":{"tool":"read_evidence"},"branch_id":"alpha","disposition":"executed","model_round":1,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":1},{"action":{"tool":"read_evidence"},"branch_id":"beta","disposition":"executed","model_round":1,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":2},{"action":{"tool":"read_evidence"},"branch_id":"gamma","disposition":"executed","model_round":1,"observation":{"error_code":null,"fact_count":1,"ok":true,"state_changed":true},"step":3},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"executed","model_round":2,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":4},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":3,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":5},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":4,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":6},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":5,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":7},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":6,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":8},{"action":{"tool":"list_evidence"},"branch_id":"controller","disposition":"executed","model_round":7,"observation":{"error_code":null,"fact_count":0,"ok":true,"state_changed":true},"step":9},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":8,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":10},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":9,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":11},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":10,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":12},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":11,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":13},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":12,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":14},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":13,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":15},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":14,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":16},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":15,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":17},{"action":{"tool":"submit_result"},"branch_id":"controller","disposition":"controller_rejected","model_round":16,"observation":{"error_code":"not_in_legal_frontier","fact_count":0,"ok":false,"state_changed":false},"step":18}],"verification":{"false_evidence_count":0,"merge_conflicts":0,"merge_omissions":3,"required_fact_recall":1.0,"utility":25,"verified_success":false}}]}