List your company

Skill retrieval on ToolBench

97Precision@RShardMemo
66.8474.6782.590.33Jan 29, 2026

Evaluation results

MethodLinks
ShardMemoModel=GPT-OSS-120B, R=3, Scope=Tier-C Skills2026.01971.94
Embedding similarityModel=GPT-OSS-120B, R=32026.01881.81
Static skill libraryModel=GPT-OSS-120B, R=32026.01861.76
BM25 retrievalModel=GPT-OSS-120B, R=32026.01851.72
RecencyModel=GPT-OSS-120B, R=32026.01791.45
RandomModel=GPT-OSS-120B, R=32026.01711.58
Trace kNNModel=GPT-OSS-120B, R=32026.01681.67