{"benchmark_id":"polymath","name":"PolyMATH","parent_benchmark":null,"categories":["math","multimodal","reasoning","spatial_reasoning","vision"],"modality":"multimodal","multilingual":false,"max_score":1.0,"language":"en","description":"Polymath is a challenging multi-modal mathematical reasoning benchmark designed to evaluate the general cognitive reasoning abilities of Multi-modal Large Language Models (MLLMs). The benchmark comprises 5,000 manually collected high-quality images of cognitive textual and visual challenges across 10 distinct categories, including pattern recognition, spatial reasoning, and relative reasoning.","paper_link":"https://arxiv.org/abs/2410.14702","implementation_link":null,"verified":false,"created_at":"2026-05-07T16:53:25.825406+00:00","updated_at":"2026-07-05T18:27:58.878321+00:00","statistics":{"total_models":23,"average_score":0.5307391304347826,"min_score":0.082,"max_score":0.865,"score_stddev":0.19198395231024917,"verified_count":0,"self_reported_count":23},"child_benchmarks":[],"linked_dataset":null,"models":[{"rank":1,"model_id":"qwen3.7-max","model_name":"Qwen3.7 Max","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.865,"normalized_score":0.865,"verified":false,"self_reported":true,"self_reported_source":"https://qwen.ai/blog?id=qwen3.7","analysis_method":null,"verification_date":null,"provider_id":"novita","input_cost_per_million":1.25,"output_cost_per_million":3.75,"context_window":1000000,"announcement_date":"2026-05-19","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":"Novita","best_throughput":null,"throughput_provider":"Novita","context_provider":"Novita"},{"rank":2,"model_id":"qwen3.7-plus","model_name":"Qwen3.7-Plus","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.84,"normalized_score":0.84,"verified":false,"self_reported":true,"self_reported_source":"https://qwen.ai/blog?id=qwen3.7-plus","analysis_method":null,"verification_date":null,"provider_id":"together","input_cost_per_million":0.32,"output_cost_per_million":1.28,"context_window":1000000,"announcement_date":"2026-05-31","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":"Together","best_throughput":null,"throughput_provider":"Together","context_provider":"Together"},{"rank":3,"model_id":"qwen3.6-plus","model_name":"Qwen3.6 Plus","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.774,"normalized_score":0.774,"verified":false,"self_reported":true,"self_reported_source":"https://qwen.ai/blog?id=qwen3.6","analysis_method":null,"verification_date":null,"provider_id":"together","input_cost_per_million":0.5,"output_cost_per_million":3.0,"context_window":1000000,"announcement_date":"2026-04-02","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":"Together","best_throughput":null,"throughput_provider":"Together","context_provider":"Together"},{"rank":4,"model_id":"qwen3.5-397b-a17b","model_name":"Qwen3.5-397B-A17B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.733,"normalized_score":0.733,"verified":false,"self_reported":true,"self_reported_source":"https://qwenlm.github.io/blog/qwen3.5/","analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-02-16","param_count":397000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":5,"model_id":"qwen3.5-27b","model_name":"Qwen3.5-27B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.712,"normalized_score":0.712,"verified":false,"self_reported":true,"self_reported_source":"https://qwen.ai/blog?id=qwen3.5","analysis_method":null,"verification_date":null,"provider_id":"novita","input_cost_per_million":0.3,"output_cost_per_million":2.4,"context_window":262144,"announcement_date":"2026-02-24","param_count":27000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":"Novita","best_throughput":null,"throughput_provider":"Novita","context_provider":"Novita"},{"rank":6,"model_id":"qwen3.5-122b-a10b","model_name":"Qwen3.5-122B-A10B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.689,"normalized_score":0.689,"verified":false,"self_reported":true,"self_reported_source":"https://qwen.ai/blog?id=qwen3.5","analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-02-24","param_count":122000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":7,"model_id":"qwen3.5-35b-a3b","model_name":"Qwen3.5-35B-A3B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.644,"normalized_score":0.644,"verified":false,"self_reported":true,"self_reported_source":"https://qwen.ai/blog?id=qwen3.5","analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-02-24","param_count":35000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":8,"model_id":"qwen3-235b-a22b-thinking-2507","model_name":"Qwen3-235B-A22B-Thinking-2507","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.601,"normalized_score":0.601,"verified":false,"self_reported":true,"self_reported_source":"https://qwenlm.github.io/blog/qwen3-thinking/","analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-07-25","param_count":235000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":9,"model_id":"qwen3.5-9b","model_name":"Qwen3.5-9B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.573,"normalized_score":0.573,"verified":false,"self_reported":true,"self_reported_source":"https://huggingface.co/Qwen/Qwen3.5-9B","analysis_method":"Thinking mode","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-03-02","param_count":9000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":10,"model_id":"qwen3-next-80b-a3b-thinking","model_name":"Qwen3-Next-80B-A3B-Thinking","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.563,"normalized_score":0.563,"verified":false,"self_reported":true,"self_reported_source":"https://qwenlm.github.io/blog/qwen3-next/","analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-09-10","param_count":80000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":11,"model_id":"qwen3-vl-32b-thinking","model_name":"Qwen3 VL 32B Thinking","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.52,"normalized_score":0.52,"verified":false,"self_reported":true,"self_reported_source":null,"analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-09-22","param_count":33000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":12,"model_id":"qwen3-vl-30b-a3b-thinking","model_name":"Qwen3 VL 30B A3B Thinking","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.517,"normalized_score":0.517,"verified":false,"self_reported":true,"self_reported_source":null,"analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-09-22","param_count":31000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":13,"model_id":"qwen3.5-4b","model_name":"Qwen3.5-4B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.511,"normalized_score":0.511,"verified":false,"self_reported":true,"self_reported_source":"https://huggingface.co/Qwen/Qwen3.5-4B","analysis_method":"Thinking mode","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-03-02","param_count":4000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":14,"model_id":"qwen3-235b-a22b-instruct-2507","model_name":"Qwen3-235B-A22B-Instruct-2507","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.502,"normalized_score":0.502,"verified":false,"self_reported":true,"self_reported_source":"https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507","analysis_method":"Accuracy","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-07-22","param_count":235000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":15,"model_id":"qwen3-vl-8b-thinking","model_name":"Qwen3 VL 8B Thinking","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.475,"normalized_score":0.475,"verified":false,"self_reported":true,"self_reported_source":null,"analysis_method":null,"verification_date":null,"provider_id":"deepinfra","input_cost_per_million":0.18,"output_cost_per_million":2.09,"context_window":262144,"announcement_date":"2025-09-22","param_count":9000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":"DeepInfra","best_throughput":null,"throughput_provider":"DeepInfra","context_provider":"DeepInfra"},{"rank":16,"model_id":"qwen3-next-80b-a3b-instruct","model_name":"Qwen3-Next-80B-A3B-Instruct","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.459,"normalized_score":0.459,"verified":false,"self_reported":true,"self_reported_source":"https://qwenlm.github.io/blog/qwen3-next/","analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-09-10","param_count":80000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":17,"model_id":"qwen3-vl-4b-thinking","model_name":"Qwen3 VL 4B Thinking","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.446,"normalized_score":0.446,"verified":false,"self_reported":true,"self_reported_source":null,"analysis_method":null,"verification_date":null,"provider_id":"deepinfra","input_cost_per_million":0.1,"output_cost_per_million":1.0,"context_window":262144,"announcement_date":"2025-09-22","param_count":4000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":"DeepInfra","best_throughput":null,"throughput_provider":"DeepInfra","context_provider":"DeepInfra"},{"rank":18,"model_id":"qwen3-vl-30b-a3b-instruct","model_name":"Qwen3 VL 30B A3B Instruct","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.443,"normalized_score":0.443,"verified":false,"self_reported":true,"self_reported_source":null,"analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-09-22","param_count":31000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":19,"model_id":"qwen3-vl-32b-instruct","model_name":"Qwen3 VL 32B Instruct","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.405,"normalized_score":0.405,"verified":false,"self_reported":true,"self_reported_source":null,"analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-09-22","param_count":33000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":20,"model_id":"qwen3-vl-8b-instruct","model_name":"Qwen3 VL 8B Instruct","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.304,"normalized_score":0.304,"verified":false,"self_reported":true,"self_reported_source":null,"analysis_method":null,"verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-09-22","param_count":9000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":21,"model_id":"qwen3-vl-4b-instruct","model_name":"Qwen3 VL 4B Instruct","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.288,"normalized_score":0.288,"verified":false,"self_reported":true,"self_reported_source":null,"analysis_method":null,"verification_date":null,"provider_id":"deepinfra","input_cost_per_million":0.1,"output_cost_per_million":0.6,"context_window":262144,"announcement_date":"2025-09-22","param_count":4000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":"DeepInfra","best_throughput":null,"throughput_provider":"DeepInfra","context_provider":"DeepInfra"},{"rank":22,"model_id":"qwen3.5-2b","model_name":"Qwen3.5-2B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.261,"normalized_score":0.261,"verified":false,"self_reported":true,"self_reported_source":"https://huggingface.co/Qwen/Qwen3.5-2B","analysis_method":"Thinking mode","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-03-02","param_count":2000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":23,"model_id":"qwen3.5-0.8b","model_name":"Qwen3.5-0.8B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.082,"normalized_score":0.082,"verified":false,"self_reported":true,"self_reported_source":"https://huggingface.co/Qwen/Qwen3.5-0.8B","analysis_method":"Thinking mode","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-03-02","param_count":800000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null}]}