Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Complex reasoning on GSM8K (test)
Loading...
87.3
Accuracy
Full Repetition
9.196
29.473
49.75
70.027
Jul 2, 2026
Accuracy
Updated 22d ago
Evaluation Results
Method
Method
Links
Accuracy
Full Repetition
Avg KV Cache=471.1, Av...
2026.07
87.3
Naïve Summary
Avg KV Cache=277.4, Av...
2026.07
85.5
PARTREP
Avg KV Cache=280.4, Av...
2026.07
85.5
No Repetition
Avg KV Cache=235.5, Av...
2026.07
84.2
LLMLingua
Avg KV Cache=277.4, Av...
2026.07
84
Echo Eviction
Avg KV Cache=235.5, Av...
2026.07
76.6
H2O Eviction
Avg KV Cache=270.8, Av...
2026.07
72.9
LLMLingua Comp.
Avg KV Cache=78.6, Avg...
2026.07
12.2
Feedback
Search any
task
Search any
task