Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Long-context understanding on RULER 32K
Loading...
92.33
Accuracy
Dense Attention
54.786
64.533
74.28
84.027
Dec 12, 2025
Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
Dense Attention
Model=Llama-3.1-8B, Sp...
2025.12
92.33
Dense Attention
Model=Llama-3.1-8B, Sp...
2025.12
92.33
BLASST
Model=Qwen3-8B, Sparsi...
2025.12
92.11
BLASST
Model=Qwen3-8B, Sparsi...
2025.12
92.08
BLASST
Model=Qwen3-8B, Sparsi...
2025.12
92.07
Dense Attention
Backbone=Qwen3-8B, Att...
2025.12
91.9
Dense Attention
Model=Qwen3-8B, Sparsi...
2025.12
91.9
Dense Attention
Model=Qwen3-8B, Sparsi...
2025.12
91.9
BLASST
Model=Llama-3.1-8B, Sp...
2025.12
91.81
BLASST
Model=Llama-3.1-8B, Sp...
2025.12
91.79
BLASST
Model=Qwen3-8B, Sparsi...
2025.12
91.74
BLASST
Model=Llama-3.1-8B, Sp...
2025.12
91.67
BLASST
Model=Llama-3.1-8B, Sp...
2025.12
91.67
BLASST
Backbone=Qwen3-8B, Att...
2025.12
91.55
RocketKV
Backbone=Qwen3-8B, Att...
2025.12
87.89
Quest
Backbone=Qwen3-8B, Att...
2025.12
56.23
Feedback
Search any
task
Search any
task