List your company

Scientific Discovery on ScienceAgentBench

0.225ScoreReAct
-0.0090.051750.11250.17325Jun 2, 2026

Evaluation results

MethodLinks
ReActBackbone=o4-mini2026.060.225
LATMBackbone=o4-mini2026.060.225
ReActBackbone=DeepSeek-V3.12026.060.206
LATMBackbone=DeepSeek-V3.12026.060.186
ML-Master2Backbone=o4-mini2026.060.176
ML-Master2Backbone=DeepSeek-V3.12026.060.157
ReActBackbone=GPT-4o2026.060.147
LAMBDABackbone=o4-mini2026.060.127
LAMBDABackbone=GPT-4o2026.060.118
EvoDS-evo-8BBackbone=Qwen3-8B2026.060.118
LAMBDABackbone=DeepSeek-V3.12026.060.108
LATMBackbone=GPT-4o2026.060.108
AutoGenBackbone=o4-mini2026.060.108
EvoDS-8BBackbone=Qwen3-8B2026.060.108
AutoGenBackbone=DeepSeek-V3.12026.060.098
Data InterpreterBackbone=o4-mini2026.060.098
Data InterpreterBackbone=DeepSeek-V3.12026.060.088
AutoGenBackbone=GPT-4o2026.060.088
Code InterpreterBackbone=GPT-4o2026.060.088
Data InterpreterBackbone=GPT-4o2026.060.088
ML-Master2Backbone=GPT-4o2026.060.088
DataMind-7BBackbone=Qwen2.5-Coder 7B2026.060.01
DataMind-14BBackbone=Qwen2.5-Coder 14B2026.060.01
DeepAnalyze-8BBackbone=DeepSeek-R1-0528-Qwen3-8B2026.060