{"benchmark_id":"textvqa","benchmark_name":"TextVQA","benchmark_description":"TextVQA contains 45,336 questions on 28,408 images that require reasoning about text to answer. Introduced to benchmark VQA models' ability to read and reason about text within images, particularly for assistive technologies for visually impaired users. The dataset addresses the gap where existing VQA datasets had few text-based questions or were too small.","max_score":1.0,"categories":["multimodal","image_to_text","vision"],"modality":"multimodal","total_models":16,"entries":[{"rank":1,"model_id":"qwen2-vl-72b","model_name":"Qwen2-VL-72B-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":"2024-08-29","announcement_date":"2024-08-29","multimodal":true,"param_count":73400000000,"is_new":false},{"rank":2,"model_id":"qwen2.5-vl-7b","model_name":"Qwen2.5 VL 7B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.849,"normalized_score":0.849,"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":"2025-01-26","announcement_date":"2025-01-26","multimodal":true,"param_count":8290000000,"is_new":false},{"rank":3,"model_id":"qwen2.5-omni-7b","model_name":"Qwen2.5-Omni-7B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.844,"normalized_score":0.844,"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":"2025-03-27","announcement_date":"2025-03-27","multimodal":true,"param_count":7000000000,"is_new":false},{"rank":4,"model_id":"lfm-2.5-vl-3b","model_name":"LFM2.5-VL-3B","organization_name":"Liquid AI","organization_id":"liquid","benchmark_score":0.843,"normalized_score":0.843,"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":3123483888,"is_new":false},{"rank":5,"model_id":"deepseek-vl2","model_name":"DeepSeek VL2","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.842,"normalized_score":0.842,"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":"2024-12-13","announcement_date":"2024-12-13","multimodal":true,"param_count":27000000000,"is_new":false},{"rank":6,"model_id":"deepseek-vl2-small","model_name":"DeepSeek VL2 Small","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.834,"normalized_score":0.834,"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":"2024-12-13","announcement_date":"2024-12-13","multimodal":true,"param_count":16000000000,"is_new":false},{"rank":7,"model_id":"nova-pro","model_name":"Nova Pro","organization_name":"Amazon","organization_id":"amazon","benchmark_score":0.815,"normalized_score":0.815,"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":"2024-11-20","announcement_date":"2024-11-20","multimodal":true,"param_count":null,"is_new":false},{"rank":8,"model_id":"deepseek-vl2-tiny","model_name":"DeepSeek VL2 Tiny","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.807,"normalized_score":0.807,"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":"2024-12-13","announcement_date":"2024-12-13","multimodal":true,"param_count":3000000000,"is_new":false},{"rank":9,"model_id":"nova-lite","model_name":"Nova Lite","organization_name":"Amazon","organization_id":"amazon","benchmark_score":0.802,"normalized_score":0.802,"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":"2024-11-20","announcement_date":"2024-11-20","multimodal":true,"param_count":null,"is_new":false},{"rank":10,"model_id":"grok-1.5v","model_name":"Grok-1.5V","organization_name":"xAI","organization_id":"xai","benchmark_score":0.781,"normalized_score":0.781,"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":"2024-04-12","announcement_date":"2024-04-12","multimodal":true,"param_count":null,"is_new":false},{"rank":11,"model_id":"phi-4-multimodal-instruct","model_name":"Phi-4-multimodal-instruct","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.756,"normalized_score":0.756,"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":"2025-02-01","announcement_date":"2025-02-01","multimodal":true,"param_count":5600000000,"is_new":false},{"rank":12,"model_id":"llama-3.2-90b-instruct","model_name":"Llama 3.2 90B Instruct","organization_name":"Meta","organization_id":"meta","benchmark_score":0.735,"normalized_score":0.735,"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":"2024-09-25","announcement_date":"2024-09-25","multimodal":true,"param_count":90000000000,"is_new":false},{"rank":13,"model_id":"phi-3.5-vision-instruct","model_name":"Phi-3.5-vision-instruct","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.72,"normalized_score":0.72,"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":"2024-08-23","announcement_date":"2024-08-23","multimodal":true,"param_count":4200000000,"is_new":false},{"rank":14,"model_id":"gemma-3-12b-it","model_name":"Gemma 3 12B","organization_name":"Google","organization_id":"google","benchmark_score":0.677,"normalized_score":0.677,"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":"2025-03-12","announcement_date":"2025-03-12","multimodal":true,"param_count":12000000000,"is_new":false},{"rank":15,"model_id":"gemma-3-27b-it","model_name":"Gemma 3 27B","organization_name":"Google","organization_id":"google","benchmark_score":0.651,"normalized_score":0.651,"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":"2025-03-12","announcement_date":"2025-03-12","multimodal":true,"param_count":27000000000,"is_new":false},{"rank":16,"model_id":"gemma-3-4b-it","model_name":"Gemma 3 4B","organization_name":"Google","organization_id":"google","benchmark_score":0.578,"normalized_score":0.578,"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":"2025-03-12","announcement_date":"2025-03-12","multimodal":true,"param_count":4000000000,"is_new":false}]}