Benchgen

RealWorldQA — Results

RankModelScore
1qwen3-8-flash-next88.5
2qwen3-8-27b85.9
3lfm2-5-vl-3b73.1
4north-micro-vision-instruct62.2
5qwen3-7-plus0.869
6seed-2-1-pro0.867
7seed-2-1-turbo0.863
8qwen3-6-plus0.854
9qwen3-6-35b-a3b0.853
10qwen3-5-122b-a10b0.851
11qwen3-5-35b-a3b0.841
12qwen3-6-27b0.841
13qwen3-5-27b0.837
14qwen3-vl-235b-a22b-thinking0.813
15qwen3-vl-235b-a22b-instruct0.793
16qwen3-vl-32b-instruct0.79
17qwen3-vl-32b-thinking0.784
18qwen3-vl-30b-a3b-thinking0.774
19qwen3-vl-30b-a3b-instruct0.737
20qwen3-vl-8b-thinking0.735
21qwen3-vl-4b-thinking0.732
22qwen3-vl-8b-instruct0.715
23qwen3-vl-4b-instruct0.709
24qwen2-5-omni-7b0.703
25grok-1-50.687

RealWorldQA — Phases

1 phaseActive

Phase 1·

RealWorldQA

Aug 3, 2026
Current