{"benchmark_id":"pointgrounding","benchmark_name":"PointGrounding","benchmark_description":"PointArena is a comprehensive platform for evaluating multimodal pointing across diverse reasoning scenarios. It includes Point-Bench, a curated dataset of ~1,000 pointing tasks across five categories: Spatial (positional references), Affordance (functional part identification), Counting (attribute-based grouping), Steerable (relative pointing), and Reasoning (open-ended visual inference). The benchmark evaluates language-guided pointing capabilities in vision-language models.","max_score":1.0,"categories":["multimodal","spatial_reasoning","grounding","vision"],"modality":"multimodal","total_models":1,"entries":[{"rank":1,"model_id":"qwen2.5-omni-7b","model_name":"Qwen2.5-Omni-7B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.665,"normalized_score":0.665,"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}]}