List your company

Tool Use Task Completion on MCP Tool-Use Tasks (evaluation)

89.9Task Success Rate (TSR)TRUSTDESC
83.86885.4348788.566Apr 8, 2026

Evaluation results

MethodLinks
TRUSTDESCModel for Gen=Claude-4.5-Sonnet, Description Type=TRUSTDESCfull2026.0489.9187
TRUSTDESCModel for Gen=Gemini-3-Flash, Description Type=TRUSTDESCfull2026.0488.5184
TRUSTDESC (Average)Model for Gen=Average, Description Type=TRUSTDESCfull2026.0487.7182.5
TRUSTDESCModel for Gen=Claude-4.5-Sonnet, Description Type=TRUSTDESClite2026.0487.5182
TRUSTDESCModel for Gen=GPT-5.2, Description Type=TRUSTDESClite2026.0487181
TRUSTDESCModel for Gen=Gemini-3-Flash, Description Type=TRUSTDESClite2026.0486.5180
TRUSTDESCModel for Gen=gpt-oss-120b, Description Type=TRUSTDESCfull2026.0486.5180
TRUSTDESC (Average)Model for Gen=Average, Description Type=TRUSTDESClite2026.0486.3179.5
TRUSTDESCModel for Gen=GPT-5.2, Description Type=TRUSTDESCfull2026.0486.1179
Original DescriptionsDescription Type=original2026.0484.1175
TRUSTDESCModel for Gen=gpt-oss-120b, Description Type=TRUSTDESClite2026.0484.1175