Over 200k context on 16GB VRAM with Qwen 3.8 27B UD-IQ3_XXS

I was using UD-Q3_K_XL until now with more than 140000 context. Quality wise it's very good, very few erroneous tool calls. Then I saw many others here reporting good results with IQ3_XXS, so I gave it a try. The downside is prompt processing speed went down from 700-800 tk/s to 400 tk/s. Qual

中文全文翻译正在补齐，可先查看下方英文原文。

I was using UD-Q3_K_XL until now with more than 140000 context. Quality wise it's very good, very few erroneous tool calls. Then I saw many others here reporting good results with IQ3_XXS, so I gave it a try. The downside is prompt processing speed went down from 700-800 tk/s to 400 tk/s. Qual