Not all quantized model perform good, Go to model page's discussion, You will find my comment with MMLU score (for detailed logs)
-
unsloth/gemma-3n-E2B-it-GGUF
Image-Text-to-Text • 4B • Updated • 104k • 35 -
unsloth/Qwen3-30B-A3B-Thinking-2507-GGUF
31B • Updated • 32.5k • 49 -
unsloth/ERNIE-4.5-21B-A3B-PT-GGUF
Text Generation • 22B • Updated • 5.47k • 8 -
Intel/Qwen3-30B-A3B-Instruct-2507-gguf-q2ks-mixed-AutoRound
31B • Updated • 454 • 11