{"benchmark":{"benchmark_id":"financial-model-benchmark","deterministic_points":"487","rubric_points":"113","sub_benchmark_id":"FMB-KNOW-ACC-01","tasks_total":60,"total_points":"600"},"generated_at":"2026-07-27T07:22:05Z","runs":[{"campaign_id":"0a0176ca-8158-41a9-99df-9a31c3ef7c31","completed_at":"2026-07-18T22:01:58.928922Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"model":{"display_name":"Kimi K3","provider":"OpenRouter (moonshotai/int4)","reasoning_effort":"max","slug":"moonshotai/kimi-k3"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"439.466666666667","maximum_points":"487"},"earned_points":"550.466666666667","maximum_points":"600","modifier_adjustment_points":"-1","percentage":"91.7444","rubric":{"earned_points":"112","maximum_points":"113"}},"started_at":"2026-07-18T21:00:24.304314Z","status":"final_model_judged","total_cost_usd":"1.8926349","updated_at":"2026-07-18T22:01:58.928922Z","usage":{"actual_cost_usd":"0.5647314","completion_tokens":28331,"prompt_tokens":59952,"total_tokens":88283},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":21061,"cost_usd":"1.3279035","model_slug":"openai/gpt-5.6-sol","prompt_tokens":115071,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":136132},"level":"model_judged"}},{"campaign_id":"38aa8e3a-02a8-4e95-bacf-a5eeb2b5cd8f","completed_at":"2026-07-16T11:53:10.333795Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"model":{"display_name":"DeepSeek V4 Flash","provider":"OpenRouter (alibaba)","reasoning_effort":"xhigh","slug":"deepseek/deepseek-v4-flash"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"457.933333333333","maximum_points":"487"},"earned_points":"569.933333333333","maximum_points":"600","modifier_adjustment_points":"0","percentage":"94.9889","rubric":{"earned_points":"112","maximum_points":"113"}},"started_at":"2026-07-16T11:44:07.40266Z","status":"final_model_judged","total_cost_usd":"0.907741394","updated_at":"2026-07-16T11:53:10.333795Z","usage":{"actual_cost_usd":"0.031676394","completion_tokens":95768,"prompt_tokens":44855,"total_tokens":140623},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":21528,"cost_usd":"0.876065","model_slug":"openai/gpt-5.6-sol","prompt_tokens":46045,"provider_name":"OpenRouter (Azure)","retries":0,"total_tokens":67573},"level":"model_judged"}},{"campaign_id":"50172fe9-380a-4d7a-a41a-404a9960d69b","completed_at":"2026-07-16T09:56:40.122404Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"model":{"display_name":"MiniMax: MiniMax M3","provider":"OpenRouter (morph)","reasoning_effort":"reasoning-max-supported","slug":"minimax/minimax-m3"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"381.599999999999","maximum_points":"487"},"earned_points":"483.266666666666","maximum_points":"600","modifier_adjustment_points":"-6.333333333333","percentage":"80.5444","rubric":{"earned_points":"108","maximum_points":"113"}},"started_at":"2026-07-16T09:39:19.186274Z","status":"final_model_judged","total_cost_usd":"1.0301266","updated_at":"2026-07-16T09:56:40.122404Z","usage":{"actual_cost_usd":"0.0541416","completion_tokens":10581,"prompt_tokens":47912,"total_tokens":58493},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":24464,"cost_usd":"0.975985","model_slug":"openai/gpt-5.6-sol","prompt_tokens":48413,"provider_name":"OpenRouter (Azure)","retries":27,"total_tokens":72877},"level":"model_judged"}},{"campaign_id":"5595b9cc-49c8-44d7-9813-8674db293fc9","completed_at":"2026-07-15T21:18:02.496007Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"model":{"display_name":"Z.ai: GLM 5.2","provider":"OpenRouter (Wafer, authenticated recovery)","reasoning_effort":"xhigh","slug":"z-ai/glm-5.2"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"463.133333333333","maximum_points":"487"},"earned_points":"576.133333333333","maximum_points":"600","modifier_adjustment_points":"0","percentage":"96.0222","rubric":{"earned_points":"113","maximum_points":"113"}},"started_at":"2026-07-15T19:58:12.554963Z","status":"final_model_judged","total_cost_usd":"1.1535194","updated_at":"2026-07-15T21:18:02.496007Z","usage":{"actual_cost_usd":"0.2824744","completion_tokens":64808,"prompt_tokens":39248,"total_tokens":104056},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":21327,"cost_usd":"0.871045","model_slug":"openai/gpt-5.6-sol","prompt_tokens":46247,"provider_name":"OpenRouter (Azure)","retries":0,"total_tokens":67574},"level":"model_judged"}},{"campaign_id":"597656e2-a55e-5b79-b457-cd576849401f","completed_at":"2026-07-27T06:42:51.144214Z","execution":{"attempted":60,"invalid_json":1,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":1,"tasks_total":60,"valid":58},"model":{"display_name":"Mistral Large 2512","provider":"OpenRouter (mistral)","reasoning_effort":"native-default","slug":"mistralai/mistral-large-2512"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"378.066666666666","maximum_points":"487"},"earned_points":"465.066666666666","maximum_points":"600","modifier_adjustment_points":"-11","percentage":"77.5111","rubric":{"earned_points":"98","maximum_points":"113"}},"started_at":"2026-07-27T06:33:06.718458Z","status":"final_model_judged","total_cost_usd":"1.4106415","updated_at":"2026-07-27T06:42:51.144214Z","usage":{"actual_cost_usd":"0.039387","completion_tokens":13664,"prompt_tokens":38601,"total_tokens":52265},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":58,"completion_tokens":23437,"cost_usd":"1.3712545","model_slug":"openai/gpt-5.6-sol","prompt_tokens":109849,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":133286},"level":"model_judged"}},{"campaign_id":"5d89e353-30a2-5dc4-a347-e529eb3890cb","completed_at":"2026-07-27T06:41:43.747447Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"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":"344.6","maximum_points":"487"},"earned_points":"431.6","maximum_points":"600","modifier_adjustment_points":"-4","percentage":"71.9333","rubric":{"earned_points":"91","maximum_points":"113"}},"started_at":"2026-07-27T06:33:04.771021Z","status":"final_model_judged","total_cost_usd":"1.5273188","updated_at":"2026-07-27T06:41:43.747447Z","usage":{"actual_cost_usd":"0.0127508","completion_tokens":7061,"prompt_tokens":35510,"total_tokens":42571},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":27099,"cost_usd":"1.514568","model_slug":"openai/gpt-5.6-sol","prompt_tokens":112782,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":139881},"level":"model_judged"}},{"campaign_id":"724a3998-0e06-5bfd-a566-21873d55bb99","completed_at":"2026-07-27T06:41:09.443988Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"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":"442.133333333333","maximum_points":"487"},"earned_points":"554.133333333333","maximum_points":"600","modifier_adjustment_points":"0","percentage":"92.3556","rubric":{"earned_points":"112","maximum_points":"113"}},"started_at":"2026-07-27T06:33:04.771019Z","status":"final_model_judged","total_cost_usd":"2.2769185","updated_at":"2026-07-27T06:41:09.443988Z","usage":{"actual_cost_usd":"0.983445","completion_tokens":20683,"prompt_tokens":93274,"total_tokens":113957},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":19938,"cost_usd":"1.2934735","model_slug":"openai/gpt-5.6-sol","prompt_tokens":115549,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":135487},"level":"model_judged"}},{"campaign_id":"8d9438e2-153d-54fc-8eb4-bbe887d287c4","completed_at":"2026-07-27T06:41:41.193733Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"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":"434.8","maximum_points":"487"},"earned_points":"543.8","maximum_points":"600","modifier_adjustment_points":"0","percentage":"90.6333","rubric":{"earned_points":"109","maximum_points":"113"}},"started_at":"2026-07-27T06:33:04.771029Z","status":"final_model_judged","total_cost_usd":"1.64108","updated_at":"2026-07-27T06:41:41.193733Z","usage":{"actual_cost_usd":"0.24771","completion_tokens":9664,"prompt_tokens":41100,"total_tokens":50764},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":23638,"cost_usd":"1.39337","model_slug":"openai/gpt-5.6-sol","prompt_tokens":111512,"provider_name":"OpenRouter (Azure EU)","retries":0,"total_tokens":135150},"level":"model_judged"}},{"campaign_id":"965caa3e-1244-4271-b7e6-e0d4c570b927","completed_at":"2026-07-16T13:56:27.274993Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"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":"449.533333333333","maximum_points":"487"},"earned_points":"559.533333333333","maximum_points":"600","modifier_adjustment_points":"0","percentage":"93.2556","rubric":{"earned_points":"110","maximum_points":"113"}},"started_at":"2026-07-16T13:37:11.637909Z","status":"final_model_judged","total_cost_usd":"1.0868295","updated_at":"2026-07-16T13:56:27.274993Z","usage":{"actual_cost_usd":"0.2080145","completion_tokens":202670,"prompt_tokens":38175,"total_tokens":240845},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":21420,"cost_usd":"0.878815","model_slug":"openai/gpt-5.6-sol","prompt_tokens":47243,"provider_name":"OpenRouter (Azure)","retries":0,"total_tokens":68663},"level":"model_judged"}},{"campaign_id":"a5827a41-b455-4539-998e-3f4de9f246ee","completed_at":"2026-07-16T13:35:08.45948Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"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":"407.457142857142","maximum_points":"487"},"earned_points":"510.457142857142","maximum_points":"600","modifier_adjustment_points":"-5","percentage":"85.0762","rubric":{"earned_points":"108","maximum_points":"113"}},"started_at":"2026-07-16T13:27:53.7536Z","status":"final_model_judged","total_cost_usd":"0.964079","updated_at":"2026-07-16T13:35:08.45948Z","usage":{"actual_cost_usd":"0.074259","completion_tokens":14084,"prompt_tokens":39261,"total_tokens":53345},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":21548,"cost_usd":"0.88982","model_slug":"openai/gpt-5.6-sol","prompt_tokens":48676,"provider_name":"OpenRouter (Azure)","retries":0,"total_tokens":70224},"level":"model_judged"}},{"campaign_id":"abd01de3-5a15-4e1c-a3a5-49e9e457d725","completed_at":"2026-07-16T10:57:51.313071Z","execution":{"attempted":60,"invalid_json":2,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":58},"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":"432.6","maximum_points":"487"},"earned_points":"541.6","maximum_points":"600","modifier_adjustment_points":"0","percentage":"90.2667","rubric":{"earned_points":"109","maximum_points":"113"}},"started_at":"2026-07-16T10:48:47.417357Z","status":"final_model_judged","total_cost_usd":"1.628511","updated_at":"2026-07-16T10:57:51.313071Z","usage":{"actual_cost_usd":"0.802131","completion_tokens":80454,"prompt_tokens":52030,"total_tokens":132484},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":58,"completion_tokens":20036,"cost_usd":"0.82638","model_slug":"openai/gpt-5.6-sol","prompt_tokens":45060,"provider_name":"OpenRouter (Azure)","retries":0,"total_tokens":65096},"level":"model_judged"}},{"campaign_id":"b13613d8-55c7-46b9-8938-460ea15a9b2f","completed_at":"2026-07-16T14:29:40.518687Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"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":"454.933333333333","maximum_points":"487"},"earned_points":"567.933333333333","maximum_points":"600","modifier_adjustment_points":"0","percentage":"94.6556","rubric":{"earned_points":"113","maximum_points":"113"}},"started_at":"2026-07-16T14:13:43.727065Z","status":"final_model_judged","total_cost_usd":"1.717681","updated_at":"2026-07-16T14:29:40.518687Z","usage":{"actual_cost_usd":"0.826446","completion_tokens":235907,"prompt_tokens":43050,"total_tokens":278957},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":22046,"cost_usd":"0.891235","model_slug":"openai/gpt-5.6-sol","prompt_tokens":45971,"provider_name":"OpenRouter (Azure)","retries":0,"total_tokens":68017},"level":"model_judged"}},{"campaign_id":"bb20db15-d7f5-46ce-a877-d741657b83bc","completed_at":"2026-07-16T11:26:43.072086Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"model":{"display_name":"Claude Sonnet 5 (Max)","provider":"OpenRouter (azure)","reasoning_effort":"max","slug":"anthropic/claude-sonnet-5"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"445.933333333333","maximum_points":"487"},"earned_points":"558.933333333333","maximum_points":"600","modifier_adjustment_points":"0","percentage":"93.1556","rubric":{"earned_points":"113","maximum_points":"113"}},"started_at":"2026-07-16T11:16:27.515454Z","status":"final_model_judged","total_cost_usd":"2.161443","updated_at":"2026-07-16T11:26:43.072086Z","usage":{"actual_cost_usd":"1.337838","completion_tokens":115129,"prompt_tokens":93274,"total_tokens":208403},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":19116,"cost_usd":"0.823605","model_slug":"openai/gpt-5.6-sol","prompt_tokens":50025,"provider_name":"OpenRouter (Azure)","retries":0,"total_tokens":69141},"level":"model_judged"}},{"campaign_id":"c5f9dc7e-c31c-44ef-bbce-e6eb1c4775d0","completed_at":"2026-07-16T06:40:51.135751Z","execution":{"attempted":60,"invalid_json":0,"missing":0,"parallelism":5,"provider_failed":0,"schema_failed":0,"tasks_total":60,"valid":60},"model":{"display_name":"xAI: Grok 4.5","provider":"OpenRouter (xai)","reasoning_effort":"high","slug":"x-ai/grok-4.5"},"provisional_reasons":[],"score":{"deterministic":{"earned_points":"464.523809523809","maximum_points":"487"},"earned_points":"575.523809523809","maximum_points":"600","modifier_adjustment_points":"0","percentage":"95.9206","rubric":{"earned_points":"111","maximum_points":"113"}},"started_at":"2026-07-16T06:15:23.081353Z","status":"final_model_judged","total_cost_usd":"1.082208","updated_at":"2026-07-16T06:40:51.135751Z","usage":{"actual_cost_usd":"0.216928","completion_tokens":19446,"prompt_tokens":60110,"total_tokens":79556},"verification":{"judge":{"assessment_not_required_tasks":0,"assessment_tasks":60,"completion_tokens":21261,"cost_usd":"0.86528","model_slug":"openai/gpt-5.6-sol","prompt_tokens":45490,"provider_name":"OpenRouter (Azure)","retries":58,"total_tokens":66751},"level":"model_judged"}}],"schema_version":"fmb-live-runs-v2"}
