Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Synthetic Long-context Retrieval on RULER 8K–64K
Loading...
79.65
Average Accuracy
Full
78.142
78.5335
78.925
79.3165
Jun 30, 2026
Average Accuracy
Updated 25d ago
Evaluation Results
Method
Method
Links
Average Accuracy
Full
Backbone=LLaMA-3.1-8B-...
2026.06
79.65
RaBitQCache
Backbone=LLaMA-3.1-8B-...
2026.06
79.55
SparQ
Backbone=LLaMA-3.1-8B-...
2026.06
79.08
Quest
Backbone=LLaMA-3.1-8B-...
2026.06
78.2
Feedback
Search any
task
Search any
task