Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Open-domain Question Answering on TriviaQA (Acc. %, AUROC, ECE)
Loading...
72.99
Accuracy (%)
Qwen3 30B
52.5748
57.8749
63.175
68.4751
Feb 18, 2026
Accuracy (%)
AUROC
Expected Calibration Error (%)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (%)
AUROC
Expected Calibration Error (%)
Qwen3 30B
confidence_type=normal...
2026.02
72.99
0.859
18.1
Gemma3 12B
confidence_type=normal...
2026.02
72.04
0.814
23.1
GLM4 9B
confidence_type=normal...
2026.02
60.52
0.748
31.5
Qwen3 4B
confidence_type=normal...
2026.02
54.21
0.829
25.5
Gemma3 4B
confidence_type=normal...
2026.02
53.36
0.775
41.1
Feedback
Search any
task
Search any
task