List your company

Science Experiment Execution on ScienceWorld (test)

51.51Success RateG-Mem
12.832422.873732.91542.9563May 8, 2026

Evaluation results

MethodLinks
G-MemExecutor=GPT-5.22026.0551.51
LangMemExecutor=Gemini-3-flash2026.0550.82
A-MemExecutor=GPT-5.22026.0550
MemCompilerExecutor=Qwen-3.5-27B2026.0548.44
A-MemExecutor=Gemini-3-flash2026.0547.78
LangMemExecutor=GPT-5.22026.0546.67
MemCompilerExecutor=Qwen-2.5-14B2026.0546.67
No MemExecutor=Gemini-3-flash2026.0545.55
G-MemExecutor=Gemini-3-flash2026.0544.44
Mem-0Executor=Gemini-3-flash2026.0542.86
A-MemExecutor=Qwen-2.5-14B2026.0534.44
G-MemExecutor=Qwen-2.5-14B2026.0532.22
G-MemExecutor=Qwen-3.5-27B2026.0532.22
A-MemExecutor=Qwen-3.5-27B2026.0532.22
No MemExecutor=Qwen-2.5-14B2026.0531.11
Mem-0Executor=Qwen-2.5-14B2026.0527.78
Mem-0Executor=Qwen-3.5-27B2026.0526.67
LangMemExecutor=Qwen-3.5-27B2026.0526.67
No MemExecutor=GPT-5.22026.0525.55
MemGenExecutor=Qwen-2.5-14B2026.0524.65
Mem-0Executor=GPT-5.22026.0522.22
No MemExecutor=Qwen-3.5-27B2026.0521.11
LangMemExecutor=Qwen-2.5-14B2026.0515.56
MemGenExecutor=Qwen-3.5-27B2026.0514.32