Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Financial numerical reasoning on DocMath Simplong (test)
Loading...
70
Accuracy
MOCA-AGENT
50.24
55.37
60.5
65.63
Jun 10, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
MOCA-AGENT
Backbone=Qwen3.6-27B,...
2026.06
70
GPT-4o
Type=Proprietary LLMs
2026.06
60
GPT-o3-mini
Type=Proprietary LLMs
2026.06
59
GPT-4.5
Type=Proprietary LLMs
2026.06
59
Qwen2.5-72B
Type=Open-source LLMs
2026.06
59
GPT-o1-preview
Type=Proprietary LLMs
2026.06
56
Qwen2.5-32B
Type=Open-source LLMs
2026.06
56
Fino1-7B
Type=Open-source LLMs
2026.06
56
Fino1-14B
Type=Open-source LLMs
2026.06
55
Llama-3.3-70B
Type=Open-source LLMs
2026.06
54
DeepSeek-V3
Type=Proprietary LLMs
2026.06
53
DeepSeek-R1
Type=Proprietary LLMs
2026.06
53
DS-R1-Distill-Llama-70B
Type=Open-source LLMs
2026.06
53
Qwen3-32B
Type=Open-source LLMs
2026.06
51
Feedback
Search any
task
Search any
task