Latest writing
Which qwen Actually Fits a 16GB GPU
No qwen3.8 tag fits a 16GB card. Two Modelfile lines make it 79% faster than the default, and qwen3.5 9B with MTP runs Q8 weights at Q4 speed.
Read the postLatest writing
No qwen3.8 tag fits a 16GB card. Two Modelfile lines make it 79% faster than the default, and qwen3.5 9B with MTP runs Q8 weights at Q4 speed.
Read the post