Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Physics Reasoning on MMLU-Pro Physics
Loading...
88.5
Accuracy
SC
40.14
52.695
65.25
77.805
Jul 8, 2026
Accuracy
Updated 16d ago
Evaluation Results
Method
Method
Links
Accuracy
SC
Backbone=Qwen3-30B-Ins...
2026.07
88.5
MILES
Backbone=Qwen3-30B-Ins...
2026.07
88.5
MILES
Backbone=GPT-OSS-20B,...
2026.07
87
ZS-CoT
Backbone=Qwen3-30B-Ins...
2026.07
87
SC
Backbone=GPT-OSS-20B,...
2026.07
86.5
BoT
Backbone=Qwen3-30B-Ins...
2026.07
86
ZS-CoT
Backbone=GPT-OSS-20B,...
2026.07
85.5
MILES
Backbone=GPT-4.1-mini,...
2026.07
85
MILES
Backbone=GPT-4.1, Infe...
2026.07
84.5
BoT
Backbone=GPT-4.1, Infe...
2026.07
84
SC
Backbone=GPT-4.1-mini,...
2026.07
83.5
DC
Backbone=GPT-4.1-mini,...
2026.07
83.5
BoT
Backbone=GPT-4.1-mini,...
2026.07
83
SC
Backbone=GPT-4.1, Infe...
2026.07
83
ZS-CoT
Backbone=GPT-4.1-mini,...
2026.07
82
ZS-CoT
Backbone=GPT-4.1, Infe...
2026.07
81
DC
Backbone=GPT-OSS-20B,...
2026.07
79
DC
Backbone=GPT-4.1, Infe...
2026.07
78
DC
Backbone=Qwen3-30B-Ins...
2026.07
63.5
BoT
Backbone=GPT-OSS-20B,...
2026.07
42
Feedback
Search any
task
Search any
task