Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Long-context evaluation on RULER 64k Zero-shot transfer
Loading...
32.9
Accuracy
Still
18.028
21.889
25.75
29.611
Jun 5, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Still
K (Cache budget)=1024,...
2026.06
32.9
Still
K (Cache budget)=2048,...
2026.06
31.8
Still
K (Cache budget)=328,...
2026.06
31.7
Still
K (Cache budget)=512,...
2026.06
27.5
Still
K (Cache budget)=164,...
2026.06
27.5
KV-Distill
K (Cache budget)=512,...
2026.06
21.8
KV-Distill
K (Cache budget)=1024,...
2026.06
21
Still
K (Cache budget)=82, N...
2026.06
20.6
KV-Distill
K (Cache budget)=2048,...
2026.06
20.2
KV-Distill
K (Cache budget)=82, N...
2026.06
19.4
KV-Distill
K (Cache budget)=328,...
2026.06
19.3
KV-Distill
K (Cache budget)=164,...
2026.06
18.6
Feedback
Search any
task
Search any
task