Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Long-context Retrieval on Needle-in-a-Haystack
Loading...
100
Retrieval Accuracy
Qwen2.5-3B-Instruct (Teacher)
67.136
75.668
84.2
92.732
Dec 23, 2025
Retrieval Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Retrieval Accuracy
Qwen2.5-3B-Instruct (Teacher)
Context length (tokens...
2025.12
100
Hybrid student (25% softmax, 75% GDN)
Context length (tokens...
2025.12
100
Qwen2.5-3B-Instruct (Teacher)
Context length (tokens...
2025.12
100
Qwen2.5-3B-Instruct (Teacher)
Context length (tokens...
2025.12
100
Qwen2.5-3B-Instruct (Teacher)
Context length (tokens...
2025.12
100
Hybrid student (25% softmax, 75% GDN)
Context length (tokens...
2025.12
99.8
Hybrid student (25% softmax, 75% GDN)
Context length (tokens...
2025.12
99.8
Hybrid student (25% softmax, 75% GDN)
Context length (tokens...
2025.12
99.4
Qwen2.5-3B-Instruct (Teacher)
Context length (tokens...
2025.12
95.4
Hybrid student (25% softmax, 75% GDN)
Context length (tokens...
2025.12
68.4
Feedback
Search any
task
Search any
task