Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

General Language Capabilities on Open LLM Leaderboard (test)

83.29HellaSwag Accuracy

TDPO

79.306880.340981.37582.4091May 12, 2026
Updated 21d ago

Evaluation Results

MethodLinks
2026.05
83.2959.0461.8651.7676.5575.167.91
2026.05
83.159.0957.8145.9177.6634.4259.66
2026.05
82.7459.4760.5444.478.2139.0460.73
2026.05
82.7259.7460.8244.4778.5839.3460.95
2026.05
82.761.5457.1743.4876.9528.6558.41
2026.05
82.3959.459.1542.377.7434.8759.3
2026.05
81.9764.0864.8453.2979.6379.2270.5
2026.05
81.964.0764.9253.3679.0878.9270.38
2026.05
81.6854.9561.4847.8875.1373.2265.72
2026.05
81.3758.760.7749.8775.1475.1266.82
2026.05
81.2356.0561.748.4775.8474.266.24
2026.05
81.0357.6759.1641.7177.2632.958.29
2026.05
80.7255.5458.4143.6776.7118.4955.59
2026.05
79.4651.4561.1946.4675.6968.6763.82