Kimi Releases Open Source PerceptionBench, GPT-5.6-Sol Achieves Top Accuracy of 59.7%

Kimi has open-sourced PerceptionBench, a multimodal large model visual perception evaluation benchmark designed to decompose visual perception capabilities into 10 atomic-level abilities for independent assessment. It covers dimensions such as visual relationships, counting, attributes, depth and 3D, localization, comparison, fine-grained recognition, context integration, OCR, and hallucination recognition. This benchmark is built on model failure cases from 42 existing evaluation sets and includes 3,000 manually verified questions, each assessing a single visual capability without requiring reasoning or external knowledge. The evaluation results show that among 16 leading multimodal large models, none achieved an overall accuracy exceeding 60%. GPT-5.6-Sol ranks first with an accuracy of 59.7%, followed by Kimi K3 (58.5%), Claude-Fable-5 (57.2%), Gemini-3.1-Pro (56.2%), and GPT-5.5 (55.8%). The report indicates that visual hallucination remains the weakest capability across all models, and there is significant room for improvement in overall perception capabilities.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.