List your company

LLM Judge Agreement on MT-bench First Turn

0.97Agreement RateG4-P
0.52280.63890.7550.8711Jun 9, 2023

Evaluation results

MethodLinks
G4-PSetup=S2 (R=50%), Comparison Judge=G4-S2023.060.97662
G4-PSetup=S2 (R=50%), Comparison Judge=Claude2023.060.94582
G4-SSetup=S2 (R=50%), Comparison Judge=Author2023.060.94175
G4-PSetup=S2 (R=50%), Comparison Judge=Author2023.060.92201
G4-SSetup=S2 (R=50%), Comparison Judge=Claude2023.060.9563
HumanSetup=S2 (R=50%), Comparison Judge=Human-M2023.060.9631
ClaudeSetup=S2 (R=50%), Comparison Judge=Author2023.060.89141
AuthorSetup=S2 (R=50%), Comparison Judge=Author2023.060.8731
ClaudeSetup=S2 (R=50%), Comparison Judge=Human-M2023.060.86414
G4-PSetup=S2 (R=50%), Comparison Judge=Human2023.060.85859
G4-PSetup=S2 (R=50%), Comparison Judge=Human-M2023.060.85546
G4-SSetup=S2 (R=50%), Comparison Judge=Human2023.060.85739
G4-SSetup=S2 (R=50%), Comparison Judge=Human-M2023.060.85473
ClaudeSetup=S2 (R=50%), Comparison Judge=Human2023.060.85648
AuthorSetup=S2 (R=50%), Comparison Judge=Human2023.060.83262
HumanSetup=S1 (R=33%), Comparison Judge=Human-M2023.060.81892
HumanSetup=S2 (R=50%), Comparison Judge=Human2023.060.81479
AuthorSetup=S2 (R=50%), Comparison Judge=Human-M2023.060.7646
G4-PSetup=S1 (R=33%), Comparison Judge=G4-S2023.060.71,138
G4-PSetup=S1 (R=33%), Comparison Judge=Author2023.060.69345
AuthorSetup=S1 (R=33%), Comparison Judge=Author2023.060.6949
G4-PSetup=S1 (R=33%), Comparison Judge=Human-M2023.060.67821
G4-SSetup=S1 (R=33%), Comparison Judge=Author2023.060.67324
G4-PSetup=S1 (R=33%), Comparison Judge=Human2023.060.661,343
G4-SSetup=S1 (R=33%), Comparison Judge=Claude2023.060.661,136
AuthorSetup=S1 (R=33%), Comparison Judge=Human2023.060.65428
G4-PSetup=S1 (R=33%), Comparison Judge=Claude2023.060.631,198
HumanSetup=S1 (R=33%), Comparison Judge=Human2023.060.63721
G4-SSetup=S1 (R=33%), Comparison Judge=Human2023.060.61,280
G4-SSetup=S1 (R=33%), Comparison Judge=Human-M2023.060.6781
ClaudeSetup=S1 (R=33%), Comparison Judge=Author2023.060.58343
ClaudeSetup=S1 (R=33%), Comparison Judge=Human-M2023.060.55820
AuthorSetup=S1 (R=33%), Comparison Judge=Human-M2023.060.5593
ClaudeSetup=S1 (R=33%), Comparison Judge=Human2023.060.541,341