Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Interactive Web Shopping on MemoryArena Bundled Web Shopping
Loading...
34.67
Success Rate (SR)
MAGE
11.0932
17.2141
23.335
29.4559
Jun 4, 2026
Success Rate (SR)
Path Success (PS)
Token Count
Updated 1mo ago
Evaluation Results
Method
Method
Links
Success Rate (SR)
Path Success (PS)
Token Count
MAGE
Backend Model=Gemma4-31B
2026.06
34.67
75.89
1,550
MAGE
Backend Model=Qwen3.6-...
2026.06
30.67
72.56
2,732
Long Context
Backend Model=Gemma4-31B
2026.06
28
71.78
3,102
Long Context
Backend Model=Qwen3.6-...
2026.06
22.67
69.78
4,092
MemoRAG
Backend Model=Qwen3.6-...
2026.06
17.33
66.22
3,888
MemoRAG
Backend Model=Gemma4-31B
2026.06
13.42
62.08
2,545
MemoryOS
Backend Model=Qwen3.6-...
2026.06
12
61.44
3,835
MemoryOS
Backend Model=Gemma4-31B
2026.06
12
60.56
1,854
Feedback
Search any
task
Search any
task