Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on AIME (Accuracy, TPR, Efficiency)
Loading...
237.5
TPR (s)
ThinKV
236.616
242.583
248.55
254.517
Oct 1, 2025
TPR (s)
Accuracy
Efficiency (Intel./Watt)
Updated 23d ago
Evaluation Results
Method
Method
Links
TPR (s)
Accuracy
Efficiency (Intel./Watt)
ThinKV
Model=R1-Llama-8B, Tok...
2025.10
237.5
46.7
0.21
R-KV (ovl)
Model=R1-Llama-8B, Tok...
2025.10
240.8
40
0.17
R-KV (seq)
Model=R1-Llama-8B, Tok...
2025.10
242.6
40
0.17
ThinKV
Model=R1-Llama-8B, Tok...
2025.10
243.6
50
0.22
R-KV (ovl)
Model=R1-Llama-8B, Tok...
2025.10
246
46.7
0.2
R-KV (seq)
Model=R1-Llama-8B, Tok...
2025.10
247.8
46.7
0.2
ThinKV
Model=R1-Llama-8B, Tok...
2025.10
251
50
0.21
R-KV (ovl)
Model=R1-Llama-8B, Tok...
2025.10
253.7
50
0.2
R-KV (seq)
Model=R1-Llama-8B, Tok...
2025.10
254.2
50
0.2
FullKV
Model=R1-Llama-8B, Tok...
2025.10
259.6
50
0.2
Feedback
Search any
task
Search any
task