Benchgen

RealWorldQA — Results

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

RealWorldQA — Evaluations

1 phaseActive

No evaluations yet for this benchmark.