Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Free-form summarization on HELMET multi_lexsum
Loading...
94.9
Utilization
Still
41.444
55.322
69.2
83.078
Jun 5, 2026
Utilization
Updated 1mo ago
Evaluation Results
Method
Method
Links
Utilization
Still
Context=8k
2026.06
94.9
AM
Context=8k
2026.06
88
Still
Context=16k
2026.06
86.8
KV-Distill
Context=8k
2026.06
85.5
Still
Context=32k
2026.06
81.3
AM
Context=16k
2026.06
80.3
KV-Distill
Context=16k
2026.06
76
Still
Context=64k
2026.06
74
AM
Context=32k
2026.06
72.1
KV-Distill
Context=32k
2026.06
68.8
AM
Context=64k
2026.06
65.2
KV-Distill
Context=64k
2026.06
59.9
Still
Context=128k
2026.06
59.3
AM
Context=128k
2026.06
49.9
KV-Distill
Context=128k
2026.06
43.5
Feedback
Search any
task
Search any
task