List your company

Multiple-choice Language Understanding on MMLU (test)

74.36Pooled AccuracyQwen3 8B
43.919251.822159.72567.6279Jun 16, 2026

Evaluation results

MethodLinks
Qwen3 8Bshot=5-shot2026.0674.3676.4412.2845
Gemma 2 9Bshot=5-shot2026.0671.4272.5415.5239
Qwen3 1.7Bshot=5-shot2026.0658.2560.2711.6733.6
Gemma 3 4Bshot=5-shot2026.0658.2359.5416.8926.5
Gemma 2 2Bshot=5-shot2026.0653.2354.6115.0122.1
Qwen3 0.6Bshot=5-shot2026.0645.0946.539.7224.6