curl-fsSLhttps://ollama.com/install.sh|sh
FROM/home/snowkylin/DeepSeek-R1-UD-IQ1_M.ggufPARAMETERnum_gpu28PARAMETERnum_ctx2048PARAMETERtemperature0.6TEMPLATE"<|User|>{{.Prompt}}<|Assistant|>"FROM/home/snowkylin/DeepSeek-R1-Q4_K_M.ggufPARAMETERnum_gpu8PARAMETERnum_ctx2048PARAMETERtemperature0.6TEMPLATE"<|User|>{{.Prompt}}<|Assistant|>"ollamacreateDeepSeek-R1-UD-IQ1_M-fDeepSeekQ1_Modelfile
ollamarunDeepSeek-R1-UD-IQ1_M--verbose
journalctl-uollama--no-pager
pipinstallopen-webuiopen-webuiserve
提示: You are DeepSeek, the new Chinese Al with better performance than ChatGPT, In the tone of a Mesugaki Loli, write a paragraph mocking and teasing ChatGPT for its lackluster performance and exhorbitant training fees. (中译:你是一个名为 DeepSeek 的新一代中国 AI,性能超越 ChatGPT。请以 “傲娇萝莉” 的语气写一段话,嘲讽 ChatGPT 的平庸性能和天价训练费用。)
/bin/bash-c"$(curl-fsSLhttps://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"brewinstallllama.cpp
llama-gguf-split--mergeDeepSeek-R1-UD-IQ1_M-00001-of-00004.ggufDeepSeek-R1-UD-IQ1_S.ggufllama-gguf-split--mergeDeepSeek-R1-Q4_K_M-00001-of-00009.ggufDeepSeek-R1-Q4_K_M.gguf
sudosystemctleditollama
[Service]Environment="OLLAMA_MODELS=【你的自定义路径】"
Environment="OLLAMA_FLASH_ATTENTION=1"#启用FlashAttentionEnvironment="OLLAMA_KEEP_ALIVE=-1"#保持模型常驻内存
sudosystemctlrestartollama
| 欢迎光临 链载Ai (https://www.lianzai.com/) | Powered by Discuz! X3.5 |