List your company
ResearchFormerly WizWand

Question Answering on HotPotQA (Initial/Final Accuracy)

55.5Initial AccuracyRPT
46.24448.64751.0553.453May 20, 2026
Updated 5mo ago

Evaluation results

MethodLinks
RPTOptimizer LLM=GPT-5, Target Model=GPT-4.12026.0555.568.4
RPTOptimizer LLM=GPT-5-mini, Target Model=GPT-4.12026.0555.565.2
RPTOptimizer LLM=Gemini-3.1-Pro, Target Model=GPT-4.12026.0555.565.8
RPTOptimizer LLM=Gemini-3.1-Flash-Lite, Target Model=GPT-4.12026.0555.564.6
ACEOptimizer LLM=GPT-5, Target Model=GPT-4.12026.0555.466.6
ACEOptimizer LLM=GPT-5-mini, Target Model=GPT-4.12026.0555.465
ACEOptimizer LLM=Gemini-3.1-Pro, Target Model=GPT-4.12026.0555.465.8
ACEOptimizer LLM=Gemini-3.1-Flash-Lite, Target Model=GPT-4.12026.0555.466.6
MIPROOptimizer LLM=GPT-5, Target Model=GPT-4.12026.0548.866.8
MIPROOptimizer LLM=GPT-5-mini, Target Model=GPT-4.12026.0548.865.6
MIPROOptimizer LLM=Gemini-3.1-Pro, Target Model=GPT-4.12026.0548.865.2
MIPROOptimizer LLM=Gemini-3.1-Flash-Lite, Target Model=GPT-4.12026.0548.863.6
GEPAOptimizer LLM=GPT-5, Target Model=GPT-4.12026.0546.664.4
GEPAOptimizer LLM=GPT-5-mini, Target Model=GPT-4.12026.0546.662.4
GEPAOptimizer LLM=Gemini-3.1-Pro, Target Model=GPT-4.12026.0546.666.4
GEPAOptimizer LLM=Gemini-3.1-Flash-Lite, Target Model=GPT-4.12026.0546.664.6