Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Math & Science on GSM8K (Score)
Loading...
88.02
Score
Full Cache
-0.9208
22.1696
45.26
68.3504
Feb 24, 2025
Score
Updated 27d ago
Evaluation Results
Method
Method
Links
Score
Full Cache
Backbone=Qwen2.5-7B-In...
2025.02
88.02
ReFreeKV
Backbone=Qwen2.5-7B-In...
2025.02
88.02
ReFreeKV
Backbone=Llama3-8B-Ins...
2025.02
76.5
PyramidKV
Backbone=Llama3-8B-Ins...
2025.02
75.44
Full Cache
Backbone=Llama3-8B-Ins...
2025.02
75.28
Heavy Hitter Oracle (H2O)
Backbone=Llama3-8B-Ins...
2025.02
74.6
CAKE
Backbone=Llama3-8B-Ins...
2025.02
74.23
StreamingLLM (SLM)
Backbone=Llama3-8B-Ins...
2025.02
72.78
SnapKV
Backbone=Llama3-8B-Ins...
2025.02
70.2
Full Cache
Backbone=Mistral-7B-In...
2025.02
33.36
ReFreeKV
Backbone=Mistral-7B-In...
2025.02
31.54
Heavy Hitter Oracle (H2O)
Backbone=Mistral-7B-In...
2025.02
2.5
Feedback
Search any
task
Search any
task