Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General language understanding and reasoning on MMLU-Redux
Loading...
83.7
Accuracy
Qwen 3 14B
56.348
63.449
70.55
77.651
Jan 13, 2026
Jan 16, 2026
Jan 20, 2026
Jan 23, 2026
Jan 27, 2026
Jan 30, 2026
Feb 3, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Qwen 3 14B
shots=5-shot
2026.01
83.7
Ministral 3 14B
shots=5-shot
2026.01
82
Qwen 3 8B
shots=5-shot
2026.01
79.4
Ministral 3 8B
shots=5-shot
2026.01
79.3
Gemma 3 12B
shots=5-shot
2026.01
76.6
Qwen 3 4B
shots=5-shot
2026.01
75.9
Ministral 3 3B
shots=5-shot
2026.01
73.5
HySparse
# Shots=5-shot, Model...
2026.02
66.2
Full-Attn
# Shots=5-shot, Model...
2026.02
65.6
Gemma 3 4B
shots=5-shot
2026.01
62.6
HySparse
# Shots=5-shot, Model...
2026.02
61.6
Hybrid SWA
# Shots=5-shot, Model...
2026.02
60.8
Full-Attn
# Shots=5-shot, Model...
2026.02
59.6
Hybrid SWA
# Shots=5-shot, Model...
2026.02
57.4
Feedback
Search any
task
Search any
task