Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Information Analysis and Processing on MARS Overall (Evaluation Set)
Loading...
1.93
Average Ranking (AI GPT-5)
Ours
1.67
3.425
5.18
6.935
Nov 3, 2025
Average Ranking (AI GPT-5)
Average Ranking (Experts sampling)
Updated 11d ago
Evaluation Results
Method
Method
Links
Average Ranking (AI GPT-5)
Average Ranking (Experts sampling)
Ours
Param.=1800B
2025.11
1.93
1.33
Qwen3-235B-A
Param.=22B
2025.11
2.77
3.77
Llama-Vision
Param.=11B
2025.11
5.23
6.3
Llama-3.2-11B
Param.=11B
2025.11
5.77
6.2
Llama-4-Scout
Param.=17B
2025.11
5.77
6.4
Qwen-VL
Param.=7B
2025.11
6.18
7.6
Llama-3.2-90B
Param.=90B
2025.11
6.77
5.9
GPT-oss
Param.=20B
2025.11
6.8
6.37
LLaVA-1.5
Param.=7B
2025.11
8.03
6.33
Qwen2-VL
Param.=72B
2025.11
8.12
7.2
Qwen2.5-VL
Param.=72B
2025.11
8.43
8.33
Feedback
Search any
task
Search any
task