Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
OpenBMBopenbmb.cn · Low evidenceOpenBMB develops open-source large language models and multimodal AI systems, including the MiniCPM series for efficient on-device inference. Their models achieve state-of-the-art performance in deep reasoning and high-frame-rate video understanding with compact parameter sizes.
Parse Score