Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Knowledge-based Reasoning on Filtered CL-Bench (test)
Loading...
34.96
Accuracy (DK)
No Memory
25.6624
28.0762
30.49
32.9038
Jun 3, 2026
Accuracy (DK)
Accuracy (ED)
Accuracy (PT)
Accuracy (RS)
Total Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (DK)
Accuracy (ED)
Accuracy (PT)
Accuracy (RS)
Total Accuracy
No Memory
Model=Qwen3-4B
2026.06
34.96
10
25.71
36.36
30.45
No Memory
Model=Qwen3-8B
2026.06
34.96
6.67
35.71
34.85
32.18
TMEM
Model=Qwen3-4B
2026.06
34.96
16.67
34.29
34.85
32.87
TMEM
Model=Qwen3-8B
2026.06
34.15
10
42.86
34.85
33.91
MemAgent
Model=Qwen3-8B
2026.06
33.33
6.67
38.57
34.85
32.18
AMEM
Model=Qwen3-8B
2026.06
31.71
10
30
30.3
28.72
AMEM
Model=Qwen3-4B
2026.06
28.46
10
25.71
33.33
26.99
MemAgent
Model=Qwen3-4B
2026.06
26.02
10
30
33.33
26.99
Feedback
Search any
task
Search any
task
