Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Science Question Answering on GPQA (Accuracy and Majority Voting)

34.8Accuracy

RLTR

21.196824.728428.2631.7916Feb 9, 2026Feb 10, 2026Feb 12, 2026Feb 14, 2026Feb 16, 2026Feb 18, 2026Feb 20, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.02
34.837.337.537.7
2026.02
3335.236.837
2026.02
32.83---
2026.02
32.435.33535.2
2026.02
32.32---
2026.02
32.32---
2026.02
31.82---
2026.02
31.82---
2026.02
30.3---
2026.02
30.3---
2026.02
30.3---
2026.02
29.8---
2026.02
29.8---
2026.02
29.8---
2026.02
29.8---
2026.02
29.8---
2026.02
29.29---
2026.02
29.29---
2026.02
29.29---
2026.02
29.29---
2026.02
28.28---
2026.02
27.78---
2026.02
27.78---
2026.02
27.78---
2026.02
25.76---
2026.02
24.24---
2026.02
24.24---
2026.02
21.72---