{"benchmark_id":"charxiv-d","benchmark_name":"CharXiv-D","benchmark_description":"CharXiv-D is the descriptive questions subset of the CharXiv benchmark, designed to assess multimodal large language models' ability to extract basic information from scientific charts. It contains descriptive questions covering information extraction, enumeration, pattern recognition, and counting across 2,323 diverse charts from arXiv papers, all curated and verified by human experts.","max_score":1.0,"categories":["multimodal","reasoning","structured_output","vision"],"modality":"multimodal","total_models":17,"entries":[{"rank":1,"model_id":"seed-2.1-pro","model_name":"Seed 2.1 Pro","organization_name":"ByteDance","organization_id":"bytedance","benchmark_score":0.955,"normalized_score":0.955,"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":"2026-06-24","announcement_date":"2026-06-24","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"seed-2.1-turbo","model_name":"Seed 2.1 Turbo","organization_name":"ByteDance","organization_id":"bytedance","benchmark_score":0.946,"normalized_score":0.946,"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":"2026-06-24","announcement_date":"2026-06-24","multimodal":true,"param_count":null,"is_new":false},{"rank":3,"model_id":"qwen3-vl-32b-instruct","model_name":"Qwen3 VL 32B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.905,"normalized_score":0.905,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":33000000000,"is_new":false},{"rank":4,"model_id":"qwen3-vl-32b-thinking","model_name":"Qwen3 VL 32B Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.902,"normalized_score":0.902,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":33000000000,"is_new":false},{"rank":5,"model_id":"gpt-4.5","model_name":"GPT-4.5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.9,"normalized_score":0.9,"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":"2025-02-27","announcement_date":"2025-02-27","multimodal":true,"param_count":null,"is_new":false},{"rank":6,"model_id":"gpt-4.1-mini-2025-04-14","model_name":"GPT-4.1 mini","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.884,"normalized_score":0.884,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.4,"output_cost_per_million":1.6,"speed_rps":150.0,"context_window":1047576,"release_date":"2025-04-14","announcement_date":"2025-04-14","multimodal":true,"param_count":null,"is_new":false},{"rank":7,"model_id":"command-a-plus-05-2026","model_name":"Command A+","organization_name":"Cohere","organization_id":"cohere","benchmark_score":0.88,"normalized_score":0.88,"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":"2026-05-20","announcement_date":"2026-05-20","multimodal":true,"param_count":218000000000,"is_new":false},{"rank":8,"model_id":"gpt-4.1-2025-04-14","model_name":"GPT-4.1","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.879,"normalized_score":0.879,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.0,"output_cost_per_million":8.0,"speed_rps":100.0,"context_window":1047576,"release_date":"2025-04-14","announcement_date":"2025-04-14","multimodal":true,"param_count":null,"is_new":false},{"rank":9,"model_id":"qwen3-vl-30b-a3b-thinking","model_name":"Qwen3 VL 30B A3B Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.869,"normalized_score":0.869,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":31000000000,"is_new":false},{"rank":10,"model_id":"qwen3-vl-8b-thinking","model_name":"Qwen3 VL 8B Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.859,"normalized_score":0.859,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":9000000000,"is_new":false},{"rank":11,"model_id":"qwen3-vl-30b-a3b-instruct","model_name":"Qwen3 VL 30B A3B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.855,"normalized_score":0.855,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":31000000000,"is_new":false},{"rank":12,"model_id":"gpt-4o-2024-08-06","model_name":"GPT-4o","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.853,"normalized_score":0.853,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.5,"output_cost_per_million":10.0,"speed_rps":132.0,"context_window":128000,"release_date":"2024-08-06","announcement_date":"2024-08-06","multimodal":true,"param_count":null,"is_new":false},{"rank":13,"model_id":"qwen3-vl-4b-thinking","model_name":"Qwen3 VL 4B Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.839,"normalized_score":0.839,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.1,"output_cost_per_million":1.0,"speed_rps":null,"context_window":262144,"release_date":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":4000000000,"is_new":false},{"rank":14,"model_id":"qwen3-vl-8b-instruct","model_name":"Qwen3 VL 8B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.83,"normalized_score":0.83,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":9000000000,"is_new":false},{"rank":15,"model_id":"qwen3-vl-4b-instruct","model_name":"Qwen3 VL 4B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.762,"normalized_score":0.762,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.1,"output_cost_per_million":0.6,"speed_rps":null,"context_window":262144,"release_date":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":4000000000,"is_new":false},{"rank":16,"model_id":"gpt-4.1-nano-2025-04-14","model_name":"GPT-4.1 nano","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.739,"normalized_score":0.739,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.1,"output_cost_per_million":0.4,"speed_rps":200.0,"context_window":1047576,"release_date":"2025-04-14","announcement_date":"2025-04-14","multimodal":true,"param_count":null,"is_new":false},{"rank":17,"model_id":"north-micro-vision-instruct","model_name":"North Micro Vision Instruct","organization_name":"Cohere","organization_id":"cohere","benchmark_score":0.6,"normalized_score":0.6,"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":"2026-08-12","announcement_date":"2026-08-12","multimodal":true,"param_count":2400000000,"is_new":false}]}