Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Deep Search on MTA (test)
Loading...
29.78
Accuracy
MTA-DeepSearch-32B
5.236
11.608
17.98
24.352
Apr 7, 2026
Accuracy
Updated 9d ago
Evaluation Results
Method
Method
Links
Accuracy
MTA-DeepSearch-32B
Evaluation Protocol=Ou...
2026.04
29.78
Gemini-3 Pro
Evaluation Protocol=Ag...
2026.04
28.65
Gemini-2.5 Pro
Evaluation Protocol=Ag...
2026.04
27.53
Gemini-3-pro
Evaluation Protocol=Di...
2026.04
26.97
GPT-5
Evaluation Protocol=Ag...
2026.04
25.84
Gemini-2.5-pro
Evaluation Protocol=Di...
2026.04
21.91
GPT-5
Evaluation Protocol=Di...
2026.04
21.35
MTA-DeepSearch-8B
Evaluation Protocol=Ou...
2026.04
20.79
Qwen3-VL-32B-Inst.
Evaluation Protocol=Ag...
2026.04
17.42
Qwen3-VL-8B-Inst.
Evaluation Protocol=Ag...
2026.04
11.8
Qwen3-VL-32B-Inst.
Evaluation Protocol=Di...
2026.04
11.24
Qwen3-VL-8B-Inst.
Evaluation Protocol=Di...
2026.04
6.18
Feedback
Search any
task
Search any
task