{"benchmark_id":"gdp-pdf","benchmark_name":"GDP.pdf","benchmark_description":"GDP.pdf is a knowledge-work vision benchmark that evaluates models on economically valuable professional tasks presented as visual documents (PDFs), testing document-based reasoning, chart and table interpretation, and problem solving without tools.","max_score":1.0,"categories":["multimodal","reasoning","general","vision"],"modality":"multimodal","total_models":11,"entries":[{"rank":1,"model_id":"claude-sonnet-5","model_name":"Claude Sonnet 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.816,"normalized_score":0.816,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":2.0,"output_cost_per_million":10.0,"speed_rps":42.0,"context_window":1000000,"release_date":"2026-06-30","announcement_date":"2026-06-30","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"gemini-3.8-flash","model_name":"Gemini 3.8 Flash","organization_name":"Google","organization_id":"google","benchmark_score":0.35,"normalized_score":0.35,"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":3,"model_id":"gemini-3.7-flash","model_name":"Gemini 3.7 Flash","organization_name":"Google","organization_id":"google","benchmark_score":0.34,"normalized_score":0.34,"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":4,"model_id":"muse-wonder-full","model_name":"Muse Wonder Full","organization_name":"Meta","organization_id":"meta","benchmark_score":0.332,"normalized_score":0.332,"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":"2026-10-12","announcement_date":"2026-10-12","multimodal":true,"param_count":null,"is_new":true},{"rank":5,"model_id":"gpt-6.1-sol","model_name":"GPT-6.1 Sol","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.32,"normalized_score":0.32,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.0,"output_cost_per_million":10.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-09-29","announcement_date":"2026-09-29","multimodal":true,"param_count":null,"is_new":false},{"rank":6,"model_id":"muse-wonder","model_name":"Muse Wonder","organization_name":"Meta","organization_id":"meta","benchmark_score":0.308,"normalized_score":0.308,"verified":false,"self_reported":true,"provider_id":"meta","input_cost_per_million":4.0,"output_cost_per_million":20.0,"speed_rps":null,"context_window":1048576,"release_date":"2026-10-12","announcement_date":"2026-10-12","multimodal":true,"param_count":null,"is_new":true},{"rank":7,"model_id":"gpt-5.6-sol","model_name":"GPT-5.6 Sol","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.307,"normalized_score":0.307,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":5.0,"output_cost_per_million":30.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":8,"model_id":"claude-fable-5","model_name":"Claude Fable 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.298,"normalized_score":0.298,"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":"2026-06-09","announcement_date":"2026-06-09","multimodal":true,"param_count":null,"is_new":false},{"rank":9,"model_id":"gpt-5.6-terra","model_name":"GPT-5.6 Terra","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.247,"normalized_score":0.247,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.0,"output_cost_per_million":12.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":10,"model_id":"gpt-5.6-luna","model_name":"GPT-5.6 Luna","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.227,"normalized_score":0.227,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.2,"output_cost_per_million":1.2,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":11,"model_id":"mistral-large-4","model_name":"Mistral Large 4","organization_name":"Mistral AI","organization_id":"mistral","benchmark_score":0.186,"normalized_score":0.186,"verified":false,"self_reported":true,"provider_id":"mistral","input_cost_per_million":0.68,"output_cost_per_million":2.09,"speed_rps":null,"context_window":1000000,"release_date":"2026-10-06","announcement_date":"2026-10-06","multimodal":true,"param_count":1050000000000,"is_new":false}]}