List your company
ResearchFormerly WizWand

Multimodal Reasoning on ZeroBench sub

47.6Pass@1Seed2.0 Pro
6.239216.977127.71538.4529May 11, 2025Jul 19, 2025Sep 26, 2025Dec 4, 2025Feb 11, 2026Apr 21, 2026Jun 30, 2026
Updated 4mo ago

Evaluation results

MethodLinks
Seed2.0 Pro2026.0647.6
Gemini-3-Pro High2026.0642.2
Seed2.0 Lite2026.0642.2
GPT-5.2 High2026.0638.9
Seed1.82026.0637.7
Seed2.0 Mini2026.0636.2
Seed 1.5-VLthinking=true, decoding=greedy2025.0530.8
Claude-Opus-4.52026.0630.8
Seed 1.5-VLthinking=false, decoding=greedy2025.0529
Gemini 1.5 Prothinking=true, decoding=greedy2025.0526
Claude 3.7 Sonnetthinking=true, decoding=sampling2025.0520.4
OpenAI o1thinking=true, decoding=greedy2025.0520.2
GPT-4othinking=false, decoding=greedy2025.0519.6
Qwen 2.5-VL 72Bthinking=false, decoding=greedy2025.0513
LLaVA-OVScale=1.5-8B2026.0611.98
Qwen2.5-VLScale=7B2026.068.99
MOSS-Video-PreviewSFT Mode=offline2026.068.53
MOSS-Video-PreviewSFT Mode=real-time2026.067.83