{"benchmark":{"benchmark_id":"financial-model-benchmark","deterministic_points":"800","rubric_points":"200","sub_benchmark_id":"FMB-CAP-ACC-03","tasks_total":100,"total_points":"1000"},"generated_at":"2026-07-27T04:59:00Z","runs":[{"campaign_id":"03700801-ab9d-510f-918b-09631c7a5164","completed_at":"2026-07-27T04:46:58.810048Z","execution":{"attempted":100,"invalid_json":1,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":99},"model":{"display_name":"Kimi K2.6","provider":"OpenRouter (modelrun/fp4)","reasoning_effort":"reasoning-max-supported","slug":"moonshotai/kimi-k2.6"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"766.190476190476","maximum_points":"800"},"earned_points":"918.933333333333","maximum_points":"1000","modifier_adjustment_points":"-34.257142857143","percentage":"91.8933","rubric":{"earned_points":"187","maximum_points":"200"}},"started_at":"2026-07-26T19:17:57.455699Z","status":"final_model_judged","total_cost_usd":"4.8148692","updated_at":"2026-07-27T04:46:58.810048Z","usage":{"actual_cost_usd":"1.9161657","completion_tokens":547020,"prompt_tokens":91991,"total_tokens":639011},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":99,"completion_tokens":47852,"cost_usd":"2.8987035","model_slug":"openai/gpt-5.6-sol","prompt_tokens":239925,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":287777},"level":"model_judged"}},{"campaign_id":"08e3a01c-cb9b-5154-a64d-c54cdbce73ca","completed_at":"2026-07-27T04:42:40.439906Z","execution":{"attempted":100,"invalid_json":1,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":99},"model":{"display_name":"Claude Sonnet 5 (Max)","provider":"OpenRouter (azure)","reasoning_effort":"max","slug":"anthropic/claude-sonnet-5"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"698.24126984127","maximum_points":"800"},"earned_points":"809.561904761904","maximum_points":"1000","modifier_adjustment_points":"-55.679365079366","percentage":"80.9562","rubric":{"earned_points":"167","maximum_points":"200"}},"started_at":"2026-07-26T19:17:49.036114Z","status":"final_model_judged","total_cost_usd":"7.8106275","updated_at":"2026-07-27T04:42:40.439906Z","usage":{"actual_cost_usd":"5.061018","completion_tokens":466377,"prompt_tokens":198624,"total_tokens":665001},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":99,"completion_tokens":42825,"cost_usd":"2.7496095","model_slug":"openai/gpt-5.6-sol","prompt_tokens":242979,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":285804},"level":"model_judged"}},{"campaign_id":"16f16b14-e315-5e40-bf9c-05c964d9f22c","completed_at":"2026-07-27T04:40:09.67841Z","execution":{"attempted":100,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":100},"model":{"display_name":"MiniMax M3","provider":"OpenRouter (morph)","reasoning_effort":"reasoning-max-supported","slug":"minimax/minimax-m3"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"466.137373737371","maximum_points":"800"},"earned_points":"273.2","maximum_points":"1000","modifier_adjustment_points":"-277.437373737371","percentage":"27.3200","rubric":{"earned_points":"84.5","maximum_points":"200"}},"started_at":"2026-07-26T19:17:45.5553Z","status":"final_model_judged","total_cost_usd":"3.64533795","updated_at":"2026-07-27T04:40:09.67841Z","usage":{"actual_cost_usd":"0.1118052","completion_tokens":21317,"prompt_tokens":101074,"total_tokens":122391},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":100,"completion_tokens":65677,"cost_usd":"3.53353275","model_slug":"openai/gpt-5.6-sol","prompt_tokens":247629,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":313306},"level":"model_judged"}},{"campaign_id":"227557ef-c8b4-50e6-bd99-57c06f7a0b78","completed_at":"2026-07-27T04:49:47.881273Z","execution":{"attempted":100,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":100},"model":{"display_name":"Claude Opus 5","provider":"OpenRouter (anthropic)","reasoning_effort":"reasoning-max-supported","slug":"anthropic/claude-opus-5"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"763.533333333334","maximum_points":"800"},"earned_points":"897.238095238096","maximum_points":"1000","modifier_adjustment_points":"-42.295238095238","percentage":"89.7238","rubric":{"earned_points":"176","maximum_points":"200"}},"started_at":"2026-07-26T19:27:42.841252Z","status":"final_model_judged","total_cost_usd":"5.1236895","updated_at":"2026-07-27T04:49:47.881273Z","usage":{"actual_cost_usd":"2.27497","completion_tokens":51274,"prompt_tokens":198624,"total_tokens":249898},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":100,"completion_tokens":45161,"cost_usd":"2.8487195","model_slug":"openai/gpt-5.6-sol","prompt_tokens":246983,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":292144},"level":"model_judged"}},{"campaign_id":"319d52a2-ddee-5abe-8fe5-8ffd7124ce35","completed_at":"2026-07-27T04:52:36.853695Z","execution":{"attempted":100,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":100},"model":{"display_name":"GPT-5.6 Terra","provider":"OpenRouter (azure)","reasoning_effort":"reasoning-max-supported","slug":"openai/gpt-5.6-terra"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"754.95873015873","maximum_points":"800"},"earned_points":"894.733333333333","maximum_points":"1000","modifier_adjustment_points":"-47.225396825397","percentage":"89.4733","rubric":{"earned_points":"187","maximum_points":"200"}},"started_at":"2026-07-26T19:27:42.841273Z","status":"final_model_judged","total_cost_usd":"3.696232","updated_at":"2026-07-27T04:52:36.853695Z","usage":{"actual_cost_usd":"0.65108","completion_tokens":28000,"prompt_tokens":92432,"total_tokens":120432},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":100,"completion_tokens":52393,"cost_usd":"3.045152","model_slug":"openai/gpt-5.6-sol","prompt_tokens":239306,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":291699},"level":"model_judged"}},{"campaign_id":"367b3b25-259f-5466-8e2f-1b8faa2b4365","completed_at":"2026-07-27T04:39:10.372733Z","execution":{"attempted":100,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":100},"model":{"display_name":"Grok 4.5","provider":"OpenRouter (xai)","reasoning_effort":"high","slug":"x-ai/grok-4.5"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"710.199999999998","maximum_points":"800"},"earned_points":"742.42380952381","maximum_points":"1000","modifier_adjustment_points":"-130.276190476188","percentage":"74.2424","rubric":{"earned_points":"162.5","maximum_points":"200"}},"started_at":"2026-07-26T19:17:45.555338Z","status":"final_model_judged","total_cost_usd":"3.565390125","updated_at":"2026-07-27T04:39:10.372733Z","usage":{"actual_cost_usd":"0.48488","completion_tokens":42670,"prompt_tokens":130750,"total_tokens":173420},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":100,"completion_tokens":52868,"cost_usd":"3.080510125","model_slug":"openai/gpt-5.6-sol","prompt_tokens":242177,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":295045},"level":"model_judged"}},{"campaign_id":"68f2a44d-b81c-5236-99bc-98bec1ae9765","completed_at":"2026-07-27T04:42:14.851402Z","execution":{"attempted":100,"invalid_json":7,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":93},"model":{"display_name":"Gemini 3.5 Flash","provider":"OpenRouter (google-vertex)","reasoning_effort":"high","slug":"google/gemini-3.5-flash"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"706.596825396826","maximum_points":"800"},"earned_points":"832.6","maximum_points":"1000","modifier_adjustment_points":"-43.496825396826","percentage":"83.2600","rubric":{"earned_points":"169.5","maximum_points":"200"}},"started_at":"2026-07-26T19:17:48.145945Z","status":"final_model_judged","total_cost_usd":"4.5754905","updated_at":"2026-07-27T04:42:14.851402Z","usage":{"actual_cost_usd":"1.8435525","completion_tokens":185220,"prompt_tokens":117715,"total_tokens":302935},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":93,"completion_tokens":45663,"cost_usd":"2.731938","model_slug":"openai/gpt-5.6-sol","prompt_tokens":222738,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":268401},"level":"model_judged"}},{"campaign_id":"69c38252-e3e5-507e-ad05-4c08906518a5","completed_at":"2026-07-27T04:39:45.813223Z","execution":{"attempted":100,"invalid_json":1,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":99},"model":{"display_name":"GLM 5.2","provider":"OpenRouter (together)","reasoning_effort":"xhigh","slug":"z-ai/glm-5.2"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"498.658585858583","maximum_points":"800"},"earned_points":"309.895238095237","maximum_points":"1000","modifier_adjustment_points":"-294.763347763346","percentage":"30.9895","rubric":{"earned_points":"106","maximum_points":"200"}},"started_at":"2026-07-26T19:17:45.555324Z","status":"final_model_judged","total_cost_usd":"3.577318275","updated_at":"2026-07-27T04:39:45.813223Z","usage":{"actual_cost_usd":"0.1802034","completion_tokens":18272,"prompt_tokens":86143,"total_tokens":104415},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":99,"completion_tokens":62633,"cost_usd":"3.397114875","model_slug":"openai/gpt-5.6-sol","prompt_tokens":241345,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":303978},"level":"model_judged"}},{"campaign_id":"6e7b2030-8926-5362-adf1-b7cf7450af08","completed_at":"2026-07-27T04:49:08.004296Z","execution":{"attempted":100,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":100},"model":{"display_name":"Kimi K3","provider":"OpenRouter (moonshotai/int4)","reasoning_effort":"max","slug":"moonshotai/kimi-k3"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"749.38095238095","maximum_points":"800"},"earned_points":"886.67619047619","maximum_points":"1000","modifier_adjustment_points":"-42.70476190476","percentage":"88.6676","rubric":{"earned_points":"180","maximum_points":"200"}},"started_at":"2026-07-26T19:18:02.35983Z","status":"final_model_judged","total_cost_usd":"4.3624279","updated_at":"2026-07-27T04:49:08.004296Z","usage":{"actual_cost_usd":"1.4579604","completion_tokens":77371,"prompt_tokens":121711,"total_tokens":199082},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":100,"completion_tokens":47016,"cost_usd":"2.9044675","model_slug":"openai/gpt-5.6-sol","prompt_tokens":245989,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":293005},"level":"model_judged"}},{"campaign_id":"71460593-cb12-5fe2-9c42-4da69ce6530c","completed_at":"2026-07-27T04:46:12.784518Z","execution":{"attempted":100,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":100},"model":{"display_name":"Qwen3.6-35B-A3B","provider":"OpenRouter (akashml/fp8)","reasoning_effort":"reasoning-max-supported","slug":"qwen/qwen3.6-35b-a3b"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"702.420634920635","maximum_points":"800"},"earned_points":"721.566666666666","maximum_points":"1000","modifier_adjustment_points":"-139.853968253969","percentage":"72.1567","rubric":{"earned_points":"159","maximum_points":"200"}},"started_at":"2026-07-26T19:17:56.785677Z","status":"final_model_judged","total_cost_usd":"3.46635614","updated_at":"2026-07-27T04:46:12.784518Z","usage":{"actual_cost_usd":"0.38941414","completion_tokens":377605,"prompt_tokens":84351,"total_tokens":461956},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":100,"completion_tokens":52555,"cost_usd":"3.076942","model_slug":"openai/gpt-5.6-sol","prompt_tokens":244114,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":296669},"level":"model_judged"}},{"campaign_id":"8e674599-5745-5237-9fc8-dd23d5f624e4","completed_at":"2026-07-27T04:43:31.664684Z","execution":{"attempted":100,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":100},"model":{"display_name":"DeepSeek V4 Flash","provider":"OpenRouter (alibaba)","reasoning_effort":"xhigh","slug":"deepseek/deepseek-v4-flash"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"732.857142857143","maximum_points":"800"},"earned_points":"776.2","maximum_points":"1000","modifier_adjustment_points":"-122.657142857143","percentage":"77.6200","rubric":{"earned_points":"166","maximum_points":"200"}},"started_at":"2026-07-26T19:17:54.118081Z","status":"final_model_judged","total_cost_usd":"3.1301241504","updated_at":"2026-07-27T04:43:31.664684Z","usage":{"actual_cost_usd":"0.1274816504","completion_tokens":429062,"prompt_tokens":97942,"total_tokens":527004},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":100,"completion_tokens":50664,"cost_usd":"3.0026425","model_slug":"openai/gpt-5.6-sol","prompt_tokens":241951,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":292615},"level":"model_judged"}},{"campaign_id":"9425845c-b842-5371-bb56-9604a5c8ed04","completed_at":"2026-07-27T04:53:54.681604Z","execution":{"attempted":100,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":100},"model":{"display_name":"Llama 4 Maverick","provider":"OpenRouter (deepinfra/base)","reasoning_effort":"native-default","slug":"meta-llama/llama-4-maverick"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"448.713997113998","maximum_points":"800"},"earned_points":"273.492063492062","maximum_points":"1000","modifier_adjustment_points":"-240.221933621936","percentage":"27.3492","rubric":{"earned_points":"65","maximum_points":"200"}},"started_at":"2026-07-26T19:31:02.975835Z","status":"final_model_judged","total_cost_usd":"3.3597306","updated_at":"2026-07-27T04:53:54.681604Z","usage":{"actual_cost_usd":"0.0249376","completion_tokens":11148,"prompt_tokens":80096,"total_tokens":91244},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":100,"completion_tokens":61377,"cost_usd":"3.334793","model_slug":"openai/gpt-5.6-sol","prompt_tokens":238064,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":299441},"level":"model_judged"}},{"campaign_id":"d433e863-5ea2-56fb-8542-40393b5aa593","completed_at":"2026-07-27T04:46:06.131058Z","execution":{"attempted":100,"invalid_json":9,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":91},"model":{"display_name":"NVIDIA Nemotron 3 Ultra","provider":"OpenRouter (together)","reasoning_effort":"high","slug":"nvidia/nemotron-3-ultra-550b-a55b"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"491.131890331888","maximum_points":"800"},"earned_points":"372.171428571429","maximum_points":"1000","modifier_adjustment_points":"-240.460461760459","percentage":"37.2171","rubric":{"earned_points":"121.5","maximum_points":"200"}},"started_at":"2026-07-26T19:17:54.118278Z","status":"final_model_judged","total_cost_usd":"3.4219172","updated_at":"2026-07-27T04:46:06.131058Z","usage":{"actual_cost_usd":"0.2577342","completion_tokens":56914,"prompt_tokens":88073,"total_tokens":144987},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":91,"completion_tokens":57762,"cost_usd":"3.164183","model_slug":"openai/gpt-5.6-sol","prompt_tokens":228734,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":286496},"level":"model_judged"}},{"campaign_id":"dbe47734-1b0c-5687-87ca-9e81f03a462a","completed_at":"2026-07-27T04:51:10.487623Z","execution":{"attempted":100,"invalid_json":3,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":100,"valid":97},"model":{"display_name":"Mistral Large 2512","provider":"OpenRouter (mistral)","reasoning_effort":"native-default","slug":"mistralai/mistral-large-2512"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"462.454834054835","maximum_points":"800"},"earned_points":"264.638095238095","maximum_points":"1000","modifier_adjustment_points":"-272.81673881674","percentage":"26.4638","rubric":{"earned_points":"75","maximum_points":"200"}},"started_at":"2026-07-26T19:58:24.355992Z","status":"final_model_judged","total_cost_usd":"3.4076145","updated_at":"2026-07-27T04:51:10.487623Z","usage":{"actual_cost_usd":"0.0919725","completion_tokens":32960,"prompt_tokens":86973,"total_tokens":119933},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":97,"completion_tokens":59877,"cost_usd":"3.315642","model_slug":"openai/gpt-5.6-sol","prompt_tokens":243034,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":302911},"level":"model_judged"}}],"schema_version":"fmb-live-runs-v2"}
