{"benchmark_id":"finance-agent-v2","benchmark_name":"Finance Agent v2","benchmark_description":"Finance Agent v2 is an agentic financial-analysis benchmark from Vals that evaluates models on real-world finance workflows, measuring their ability to retrieve and reason over financial documents, perform multi-step calculations, and produce accurate analyses.","max_score":1.0,"categories":["reasoning","finance","agents"],"modality":"text","total_models":26,"entries":[{"rank":1,"model_id":"gemini-3.5-flash","model_name":"Gemini 3.5 Flash","organization_name":"Google","organization_id":"google","benchmark_score":0.5786,"normalized_score":0.5786,"verified":false,"self_reported":false,"provider_id":"google","input_cost_per_million":1.5,"output_cost_per_million":9.0,"speed_rps":null,"context_window":1048576,"release_date":"2026-05-19","announcement_date":"2026-05-19","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"muse-spark-1.1","model_name":"Muse Spark 1.1","organization_name":"Meta","organization_id":"meta","benchmark_score":0.572,"normalized_score":0.572,"verified":false,"self_reported":true,"provider_id":"meta","input_cost_per_million":1.25,"output_cost_per_million":4.25,"speed_rps":null,"context_window":1048576,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":3,"model_id":"claude-fable-5","model_name":"Claude Fable 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.5631,"normalized_score":0.5631,"verified":false,"self_reported":false,"provider_id":"anthropic","input_cost_per_million":10.0,"output_cost_per_million":50.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-06-09","announcement_date":"2026-06-09","multimodal":true,"param_count":null,"is_new":false},{"rank":4,"model_id":"claude-opus-4-8","model_name":"Claude Opus 4.8","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.5392,"normalized_score":0.5392,"verified":false,"self_reported":false,"provider_id":"anthropic","input_cost_per_million":5.0,"output_cost_per_million":25.0,"speed_rps":42.0,"context_window":1000000,"release_date":"2026-05-28","announcement_date":"2026-05-28","multimodal":true,"param_count":null,"is_new":false},{"rank":5,"model_id":"gpt-5.5","model_name":"GPT-5.5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.5176,"normalized_score":0.5176,"verified":false,"self_reported":false,"provider_id":"openai","input_cost_per_million":5.0,"output_cost_per_million":30.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-04-23","announcement_date":"2026-04-23","multimodal":true,"param_count":null,"is_new":false},{"rank":6,"model_id":"claude-opus-4-7","model_name":"Claude Opus 4.7","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.5151,"normalized_score":0.5151,"verified":false,"self_reported":false,"provider_id":"anthropic","input_cost_per_million":5.0,"output_cost_per_million":25.0,"speed_rps":42.0,"context_window":1000000,"release_date":"2026-04-16","announcement_date":"2026-04-16","multimodal":true,"param_count":null,"is_new":false},{"rank":7,"model_id":"claude-sonnet-4-6","model_name":"Claude Sonnet 4.6","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.5103,"normalized_score":0.5103,"verified":false,"self_reported":false,"provider_id":"anthropic","input_cost_per_million":3.0,"output_cost_per_million":15.0,"speed_rps":42.0,"context_window":200000,"release_date":"2026-02-17","announcement_date":"2026-02-17","multimodal":true,"param_count":null,"is_new":false},{"rank":8,"model_id":"qwen3.7-max","model_name":"Qwen3.7 Max","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.4835,"normalized_score":0.4835,"verified":false,"self_reported":false,"provider_id":"novita","input_cost_per_million":1.25,"output_cost_per_million":3.75,"speed_rps":null,"context_window":1000000,"release_date":"2026-05-19","announcement_date":"2026-05-19","multimodal":false,"param_count":null,"is_new":false},{"rank":9,"model_id":"minimax-m3","model_name":"MiniMax M3","organization_name":"MiniMax","organization_id":"minimax","benchmark_score":0.4827,"normalized_score":0.4827,"verified":false,"self_reported":false,"provider_id":"novita","input_cost_per_million":0.3,"output_cost_per_million":1.2,"speed_rps":null,"context_window":1000000,"release_date":"2026-06-01","announcement_date":"2026-06-01","multimodal":true,"param_count":428000000000,"is_new":false},{"rank":10,"model_id":"gpt-5.4-mini","model_name":"GPT-5.4 mini","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.4536,"normalized_score":0.4536,"verified":false,"self_reported":false,"provider_id":"openai","input_cost_per_million":0.75,"output_cost_per_million":4.5,"speed_rps":null,"context_window":400000,"release_date":"2026-03-17","announcement_date":"2026-03-17","multimodal":true,"param_count":null,"is_new":false},{"rank":11,"model_id":"kimi-k2.6","model_name":"Kimi K2.6","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.4487,"normalized_score":0.4487,"verified":false,"self_reported":false,"provider_id":"deepinfra","input_cost_per_million":0.75,"output_cost_per_million":3.5,"speed_rps":null,"context_window":262144,"release_date":"2026-04-20","announcement_date":"2026-04-20","multimodal":true,"param_count":1000000000000,"is_new":false},{"rank":12,"model_id":"glm-5.1","model_name":"GLM-5.1","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.4479,"normalized_score":0.4479,"verified":false,"self_reported":false,"provider_id":"friendli","input_cost_per_million":1.4,"output_cost_per_million":4.4,"speed_rps":null,"context_window":200000,"release_date":"2026-04-07","announcement_date":"2026-04-07","multimodal":false,"param_count":754000000000,"is_new":false},{"rank":13,"model_id":"gemini-3.1-pro-preview","model_name":"Gemini 3.1 Pro","organization_name":"Google","organization_id":"google","benchmark_score":0.4298,"normalized_score":0.4298,"verified":false,"self_reported":false,"provider_id":"google","input_cost_per_million":2.5,"output_cost_per_million":15.0,"speed_rps":90.0,"context_window":1048576,"release_date":"2026-02-19","announcement_date":"2026-02-19","multimodal":true,"param_count":null,"is_new":false},{"rank":14,"model_id":"gemini-3-flash-preview","model_name":"Gemini 3 Flash","organization_name":"Google","organization_id":"google","benchmark_score":0.4255,"normalized_score":0.4255,"verified":false,"self_reported":false,"provider_id":"google","input_cost_per_million":0.5,"output_cost_per_million":3.0,"speed_rps":null,"context_window":1000000,"release_date":"2025-12-17","announcement_date":"2025-12-17","multimodal":true,"param_count":null,"is_new":false},{"rank":15,"model_id":"mimo-v2.5-pro","model_name":"MiMo-V2.5-Pro","organization_name":"Xiaomi","organization_id":"xiaomi","benchmark_score":0.415,"normalized_score":0.415,"verified":false,"self_reported":false,"provider_id":"xiaomi","input_cost_per_million":0.435,"output_cost_per_million":0.87,"speed_rps":null,"context_window":1048576,"release_date":"2026-04-27","announcement_date":"2026-04-27","multimodal":false,"param_count":1023244718976,"is_new":false},{"rank":16,"model_id":"qwen3.6-plus","model_name":"Qwen3.6 Plus","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.4085,"normalized_score":0.4085,"verified":false,"self_reported":false,"provider_id":"together","input_cost_per_million":0.5,"output_cost_per_million":3.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-03-31","announcement_date":"2026-04-02","multimodal":true,"param_count":null,"is_new":false},{"rank":17,"model_id":"qwen3.7-plus","model_name":"Qwen3.7-Plus","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.3822,"normalized_score":0.3822,"verified":false,"self_reported":false,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"speed_rps":null,"context_window":null,"release_date":"2026-05-31","announcement_date":"2026-05-31","multimodal":true,"param_count":null,"is_new":false},{"rank":18,"model_id":"gpt-5.4-nano","model_name":"GPT-5.4 nano","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.3822,"normalized_score":0.3822,"verified":false,"self_reported":false,"provider_id":"openai","input_cost_per_million":0.2,"output_cost_per_million":1.25,"speed_rps":null,"context_window":400000,"release_date":"2026-03-17","announcement_date":"2026-03-17","multimodal":true,"param_count":null,"is_new":false},{"rank":19,"model_id":"grok-4.3","model_name":"Grok 4.3","organization_name":"xAI","organization_id":"xai","benchmark_score":0.3771,"normalized_score":0.3771,"verified":false,"self_reported":false,"provider_id":"xai","input_cost_per_million":1.25,"output_cost_per_million":2.5,"speed_rps":null,"context_window":1000000,"release_date":"2026-05-06","announcement_date":"2026-05-06","multimodal":false,"param_count":null,"is_new":false},{"rank":20,"model_id":"nemotron-3-ultra-550b-a55b","model_name":"Nemotron 3 Ultra (550B A55B)","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.3753,"normalized_score":0.3753,"verified":false,"self_reported":false,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"speed_rps":null,"context_window":null,"release_date":"2026-06-04","announcement_date":"2026-06-04","multimodal":false,"param_count":550000000000,"is_new":false},{"rank":21,"model_id":"mimo-v2.5","model_name":"MiMo-V2.5","organization_name":"Xiaomi","organization_id":"xiaomi","benchmark_score":0.3673,"normalized_score":0.3673,"verified":false,"self_reported":false,"provider_id":"novita","input_cost_per_million":0.168,"output_cost_per_million":0.336,"speed_rps":null,"context_window":1048576,"release_date":"2026-04-22","announcement_date":"2026-04-22","multimodal":true,"param_count":310775040000,"is_new":false},{"rank":22,"model_id":"mistral-medium-3-5","model_name":"Mistral Medium 3.5","organization_name":"Mistral AI","organization_id":"mistral","benchmark_score":0.321,"normalized_score":0.321,"verified":false,"self_reported":false,"provider_id":"mistral","input_cost_per_million":1.5,"output_cost_per_million":7.5,"speed_rps":null,"context_window":256000,"release_date":"2026-04-29","announcement_date":"2026-04-29","multimodal":true,"param_count":128000000000,"is_new":false},{"rank":23,"model_id":"claude-haiku-4-5-20251001","model_name":"Claude Haiku 4.5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.3101,"normalized_score":0.3101,"verified":false,"self_reported":false,"provider_id":"anthropic","input_cost_per_million":1.0,"output_cost_per_million":5.0,"speed_rps":100.0,"context_window":200000,"release_date":"2025-10-15","announcement_date":"2025-10-15","multimodal":true,"param_count":null,"is_new":false},{"rank":24,"model_id":"gemini-3.1-flash-lite-preview","model_name":"Gemini 3.1 Flash-Lite","organization_name":"Google","organization_id":"google","benchmark_score":0.2999,"normalized_score":0.2999,"verified":false,"self_reported":false,"provider_id":"google","input_cost_per_million":0.25,"output_cost_per_million":1.5,"speed_rps":null,"context_window":1000000,"release_date":"2026-03-03","announcement_date":"2026-03-03","multimodal":true,"param_count":null,"is_new":false},{"rank":25,"model_id":"grok-4.20-beta-0309-reasoning","model_name":"Grok-4.20 Beta Reasoning","organization_name":"xAI","organization_id":"xai","benchmark_score":0.2849,"normalized_score":0.2849,"verified":false,"self_reported":false,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"speed_rps":null,"context_window":null,"release_date":"2026-03-09","announcement_date":"2026-03-09","multimodal":true,"param_count":null,"is_new":false},{"rank":26,"model_id":"minimax-m2.7","model_name":"MiniMax M2.7","organization_name":"MiniMax","organization_id":"minimax","benchmark_score":0.2789,"normalized_score":0.2789,"verified":false,"self_reported":false,"provider_id":"fireworks","input_cost_per_million":0.3,"output_cost_per_million":1.2,"speed_rps":null,"context_window":196608,"release_date":"2026-03-18","announcement_date":"2026-03-18","multimodal":false,"param_count":null,"is_new":false}]}