yall are sleeping on qwen 3.8 27b q2 + q2 dflash + q5 kv
英文标题:yall are sleeping on qwen 3.8 27b q2 + q2 dflash + q5 kv
ok bit more context: it's actually a QAT Q2 for Qwen 3.8 27 B: https://huggingface.co/sdkyuan/qwen3.8-27B-qat-q2_0-gguf QAT Q2 for DFlash model: https://huggingface.co/HermiHg/Qwen3.8-27B-DFlash2-Q2_K_S-MIX-GGUF Q5 KV seems to cause 0 problems for me; I've used it up to 200K tokens of context. Total
打开原文相关 2 · 机会 0
中文
ok bit more context: it's actually a QAT Q2 for Qwen 3.8 27 B: https://huggingface.co/sdkyuan/qwen3.8-27B-qat-q2_0-gguf QAT Q2 for DFlash model: https://huggingface.co/HermiHg/Qwen3.8-27B-DFlash2-Q2_K_S-MIX-GGUF Q5 KV seems to cause 0 problems for me; I've used it up to 200K tokens of context. Total
中文全文翻译正在补齐,可先查看下方英文原文。
English / 原文
ok bit more context: it's actually a QAT Q2 for Qwen 3.8 27 B: https://huggingface.co/sdkyuan/qwen3.8-27B-qat-q2_0-gguf QAT Q2 for DFlash model: https://huggingface.co/HermiHg/Qwen3.8-27B-DFlash2-Q2_K_S-MIX-GGUF Q5 KV seems to cause 0 problems for me; I've used it up to 200K tokens of context. Total