Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Table Reasoning on MMTU (test)
Loading...
56
Accuracy
CPO
22.928
31.514
40.1
48.686
Jun 3, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
CPO
Model=Qwen3-14B-Base
2026.06
56
Table-R1
Model=Table-R1
2026.06
51
DPO
Model=Mistral-8B-Instruct
2026.06
49.8
CPO
Model=Mistral-8B-Instruct
2026.06
46
Trace SFT
Model=Qwen3-14B-Base
2026.06
44
CPO
Model=Llama-3.1-8B-Ins...
2026.06
43
Trace SFT
Model=Llama-3.1-8B-Ins...
2026.06
40
DPO
Model=Llama-3.1-8B-Ins...
2026.06
38.3
Trace SFT
Model=Mistral-8B-Instruct
2026.06
33
Q&A SFT
Model=Llama-3.1-8B-Ins...
2026.06
31
Q&A SFT
Model=Mistral-8B-Instruct
2026.06
30
Q&A SFT
Model=Qwen3-14B-Base
2026.06
29
DPO
Model=Qwen3-14B-Base
2026.06
24.2
Feedback
Search any
task
Search any
task