This draft model is OP on 16 GB cards for Qwen 3.8 27b
英文标题：This draft model is OP on 16 GB cards for Qwen 3.8 27b
https://huggingface.co/HermiHg/Qwen3.8-27B-DFlash2-Q2_K_S-MIX-GGUF I used this draft model with https://huggingface.co/ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF with the IQ3_XXS with 128k context and I saw it averaging about 60 tokens per second tg speed on the 16 GB RX 9070 XT. This is much better than
打开原文相关 2 · 机会 0
中文
https://huggingface.co/HermiHg/Qwen3.8-27B-DFlash2-Q2_K_S-MIX-GGUF I used this draft model with https://huggingface.co/ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF with the IQ3_XXS with 128k context and I saw it averaging about 60 tokens per second tg speed on the 16 GB RX 9070 XT. This is much better than
中文全文翻译正在补齐，可先查看下方英文原文。
English / 原文
https://huggingface.co/HermiHg/Qwen3.8-27B-DFlash2-Q2_K_S-MIX-GGUF I used this draft model with https://huggingface.co/ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF with the IQ3_XXS with 128k context and I saw it averaging about 60 tokens per second tg speed on the 16 GB RX 9070 XT. This is much better than