Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Tool Selection on ToolBench 30% 2,413 tools, 180 queries (test)
Loading...
39.2
Recall@1
BM25
23.184
27.342
31.5
35.658
Mar 13, 2026
Recall@1
Recall@3
Recall@5
NDCG@5
Updated 1mo ago
Evaluation Results
Method
Method
Links
Recall@1
Recall@3
Recall@5
NDCG@5
BM25
2026.03
39.2
77.2
92.1
85.3
SE + Lexical
2026.03
38.8
78.3
91.9
85.4
OATS-S3
2026.03
38.7
77.8
90.6
84.1
SE
2026.03
38.2
76.5
89.7
83.4
OATS-S1
2026.03
38.1
79.3
91.5
84.8
OATS-S2
2026.03
37.2
77.4
89.6
82.3
Random
2026.03
23.8
65.6
82.9
69.2
Feedback
Search any
task
Search any
task
