Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Pairwise Ranking on LFQA
Loading...
77.24
Pairwise Preference Accuracy
Claude-3.5 Sonnet
62.5136
66.3368
70.16
73.9832
Dec 17, 2024
Pairwise Preference Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Pairwise Preference Accuracy
Claude-3.5 Sonnet
2024.12
77.24
GPT-4o
2024.12
76.54
LMUNIT LLaMA3.1-70B-Decomposed-Weighted
Base Model Architectur...
2024.12
76.53
LMUNIT LLaMA3.1-70B
Base Model Architectur...
2024.12
76.15
SFR-LLaMA-3.1-70B-Judge
Base Model Architectur...
2024.12
75
LMUNIT LLaMA3.1-70B-Decomposed
Base Model Architectur...
2024.12
74.62
Prometheus-2-8x7B
Base Model Architectur...
2024.12
74.23
Prometheus-2-7B
Base Model Architectur...
2024.12
72.31
Prometheus-2-BGB-8x7B
2024.12
71.54
LMUNIT LLaMA3.1-8B
Base Model Architectur...
2024.12
71.54
SFR-LLaMA-3.1-8B-Judge
Base Model Architectur...
2024.12
68.85
Skywork-Critic-Llama-3.1-8B
Base Model Architectur...
2024.12
64.23
Llama-3-OffsetBias-8B
Base Model Architectur...
2024.12
63.08
Feedback
Search any
task
Search any
task