{"benchmark_id":"gdpval-aa","benchmark_name":"GDPval-AA","benchmark_description":"GDPval-AA evaluates AI agents on economically valuable professional knowledge-work tasks and reports performance as an Elo score.","max_score":3000.0,"categories":["legal","reasoning","finance","general","agents"],"modality":"text","total_models":11,"entries":[{"rank":1,"model_id":"claude-fable-5-1","model_name":"Claude Fable 5.1","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":1853.0,"normalized_score":0.618,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":10.0,"output_cost_per_million":50.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-09-01","announcement_date":"2026-09-01","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"glm-5.3-flash","model_name":"GLM-5.3-Flash","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":1773.0,"normalized_score":0.591,"verified":false,"self_reported":true,"provider_id":"z","input_cost_per_million":0.15,"output_cost_per_million":0.5,"speed_rps":null,"context_window":1000000,"release_date":"2026-08-26","announcement_date":"2026-08-26","multimodal":true,"param_count":320000000000,"is_new":false},{"rank":3,"model_id":"glm-5.3","model_name":"GLM-5.3","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":1769.0,"normalized_score":0.59,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":1.2,"output_cost_per_million":4.0,"speed_rps":null,"context_window":1048576,"release_date":"2026-08-14","announcement_date":"2026-08-14","multimodal":false,"param_count":753000000000,"is_new":false},{"rank":4,"model_id":"muse-spark-1.3","model_name":"Muse Spark 1.3","organization_name":"Meta","organization_id":"meta","benchmark_score":1754.0,"normalized_score":0.585,"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":"2026-09-02","announcement_date":"2026-09-02","multimodal":true,"param_count":null,"is_new":false},{"rank":5,"model_id":"grok-4.6","model_name":"Grok 4.6","organization_name":"xAI","organization_id":"xai","benchmark_score":1753.0,"normalized_score":0.584,"verified":false,"self_reported":true,"provider_id":"xai","input_cost_per_million":2.0,"output_cost_per_million":6.0,"speed_rps":null,"context_window":500000,"release_date":"2026-08-12","announcement_date":"2026-08-12","multimodal":true,"param_count":null,"is_new":false},{"rank":6,"model_id":"hy4-preview","model_name":"Hy4 preview","organization_name":"Tencent","organization_id":"tencent","benchmark_score":1678.0,"normalized_score":0.559,"verified":false,"self_reported":true,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"speed_rps":null,"context_window":null,"release_date":"2026-08-28","announcement_date":"2026-08-28","multimodal":false,"param_count":770000000000,"is_new":false},{"rank":7,"model_id":"gemini-3.8-flash","model_name":"Gemini 3.8 Flash","organization_name":"Google","organization_id":"google","benchmark_score":1545.0,"normalized_score":0.515,"verified":false,"self_reported":true,"provider_id":"google","input_cost_per_million":0.75,"output_cost_per_million":3.75,"speed_rps":null,"context_window":1048576,"release_date":"2026-09-02","announcement_date":"2026-09-02","multimodal":true,"param_count":null,"is_new":false},{"rank":8,"model_id":"gemini-3.7-flash","model_name":"Gemini 3.7 Flash","organization_name":"Google","organization_id":"google","benchmark_score":1525.0,"normalized_score":0.508,"verified":false,"self_reported":true,"provider_id":"google","input_cost_per_million":0.75,"output_cost_per_million":3.75,"speed_rps":null,"context_window":1048576,"release_date":"2026-08-13","announcement_date":"2026-08-13","multimodal":true,"param_count":null,"is_new":false},{"rank":9,"model_id":"inkling-small","model_name":"Inkling-Small","organization_name":"Thinking Machines Lab","organization_id":"thinking-machines","benchmark_score":1269.0,"normalized_score":0.423,"verified":false,"self_reported":true,"provider_id":"thinking-machines","input_cost_per_million":0.3,"output_cost_per_million":1.2,"speed_rps":null,"context_window":256000,"release_date":"2026-07-30","announcement_date":"2026-07-30","multimodal":true,"param_count":276000000000,"is_new":false},{"rank":10,"model_id":"inkling","model_name":"Inkling","organization_name":"Thinking Machines Lab","organization_id":"thinking-machines","benchmark_score":1238.0,"normalized_score":0.413,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.95,"output_cost_per_million":4.05,"speed_rps":null,"context_window":524288,"release_date":"2026-07-21","announcement_date":"2026-07-21","multimodal":true,"param_count":975000000000,"is_new":false},{"rank":11,"model_id":"ling-3.0-flash","model_name":"Ling 3.0 Flash","organization_name":"InclusionAI","organization_id":"inclusionai","benchmark_score":1107.0,"normalized_score":0.369,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.06,"output_cost_per_million":0.18,"speed_rps":null,"context_window":131072,"release_date":"2026-08-04","announcement_date":"2026-08-04","multimodal":false,"param_count":124000000000,"is_new":false}]}