CoolFace
Modelpublic

Alcoft/dnotitia_Smoothie-Qwen3-14B-GGUF

sourceHugging Faceapache-2.0updated 1y agoView on Hugging Face
0likes1.1kdownloads
Model Card
QuantSizeDescription
Q2_K_XXS5.0 GBNot recommended for most people. Extremelly low quality.
Q2_K_XS5.17 GBNot recommended for most people. Very low quality.
Q2_K5.36 GBNot recommended for most people. Very low quality.
Q2_K_L6.07 GBNot recommended for most people. Uses Q80 for output and embedding, and Q2K for everything else. Very low quality.
Q2_K_XL7.42 GBNot recommended for most people. Uses F16 for output and embedding, and Q2_K for everything else. Very low quality.
Q3_K_XXS5.92 GBNot recommended for most people. Prefer any bigger Q3_K quantization. Very low quality.
Q3_K_XS6.01 GBNot recommended for most people. Prefer any bigger Q3_K quantization. Very low quality.
Q3_K_S6.2 GBNot recommended for most people. Prefer any bigger Q3_K quantization. Low quality.
Q3_K_M6.82 GBNot recommended for most people. Low quality.
Q3_K_L7.36 GBNot recommended for most people. Low quality.
Q3_K_XL7.99 GBNot recommended for most people. Uses Q80 for output and embedding, and Q3K_L for everything else. Low quality.
Q3_K_XXL9.35 GBNot recommended for most people. Uses F16 for output and embedding, and Q3KL for everything else. Low quality.
Q4_K_XS7.8 GBLower quality than Q4KS.
Q4_K_S7.98 GBRecommended. Slightly low quality.
Q4_K_M8.38 GBRecommended. Decent quality for most use cases.
Q4_K_L8.92 GBRecommended. Uses Q80 for output and embedding, and Q4K_M for everything else. Decent quality.
Q4_K_XL10.28 GBRecommended. Uses F16 for output and embedding, and Q4KM for everything else. Decent quality.
Q5_K_XXS9.37 GBLower quality than Q5KS.
Q5_K_XS9.46 GBLower quality than Q5KS.
Q5_K_S9.56 GBRecommended. High quality.
Q5_K_M9.79 GBRecommended. High quality.
Q5_K_L10.24 GBRecommended. Uses Q80 for output and embedding, and Q5K_M for everything else. High quality.
Q5_K_XL11.6 GBRecommended. Uses F16 for output and embedding, and Q5KM for everything else. High quality.
Q6_K_S11.1 GBLower quality than Q6_K.
Q6_K11.29 GBRecommended. Very high quality.
Q6_K_L11.64 GBRecommended. Uses Q80 for output and embedding, and Q6K for everything else. Very high quality.
Q6_K_XL13.0 GBRecommended. Uses F16 for output and embedding, and Q6_K for everything else. Very high quality.
Q8_K_XS14.26 GBLower quality than Q8_0.
Q8_K_S14.44 GBLower quality than Q8_0.
Q8_014.62 GBRecommended. Quality almost like F16.
Q8_K_XL15.98 GBRecommended. Uses F16 for output and embedding, and Q8_0 for everything else. Quality almost like F16.
F1627.51 GBNot recommended. Overkill. Prefer Q8_0.
ORIGINAL (BF16)27.51 GBNot recommended. Overkill. Prefer Q8_0.

Quantized using TAO71-AI AutoQuantizer. You can check out the original model card here.