<![CDATA[arkohut的博客]]>https://blog.csdn.net/arkohutzh-cnhttps://blog.csdn.net/<![CDATA[Gradio Tunneling 支持固定域名啦]]>https://blog.csdn.net/arkohut/article/details/145234394https://blog.csdn.net/arkohut/article/details/145234394arkohutSat, 18 Jan 2025 23:02:01 +0800<![CDATA[健忘症福音...Rewind 平替...自动记录屏幕的 AI 应用来啦]]>https://blog.csdn.net/arkohut/article/details/143839745https://blog.csdn.net/arkohut/article/details/143839745arkohutSun, 17 Nov 2024 22:09:30 +0800<![CDATA[卷卷卷！新的开源多模态大模型王者来啦，从量化到部署用 lmdeploy 把它运行起来吧]]>https://blog.csdn.net/arkohut/article/details/140309405https://blog.csdn.net/arkohut/article/details/140309405arkohutTue, 09 Jul 2024 23:35:05 +0800<![CDATA[safetensors -＞ ollama ，模型转化 + 模型量化详细步骤]]>https://blog.csdn.net/arkohut/article/details/140087605https://blog.csdn.net/arkohut/article/details/140087605arkohutSun, 30 Jun 2024 22:18:47 +0800<![CDATA[GLM4 9B - 环境准备和 vllm 部署]]>https://blog.csdn.net/arkohut/article/details/139507027https://blog.csdn.net/arkohut/article/details/139507027arkohutThu, 06 Jun 2024 18:09:19 +0800<![CDATA[Ollama 支持的 flash attention 能提升推理速度吗？我们一起测测看吧]]>https://blog.csdn.net/arkohut/article/details/139426912https://blog.csdn.net/arkohut/article/details/139426912arkohutMon, 03 Jun 2024 23:30:28 +0800<![CDATA[ollama 支持并发之后和 vllm 相比性能如何？我们测测看]]>https://blog.csdn.net/arkohut/article/details/139076652https://blog.csdn.net/arkohut/article/details/139076652arkohutMon, 20 May 2024 22:34:47 +0800<![CDATA[大模型推理 A40 vs A6000 谁更强 - 对比 Yi-34B 的单、双卡推理性能]]>https://blog.csdn.net/arkohut/article/details/135613641https://blog.csdn.net/arkohut/article/details/135613641arkohutMon, 15 Jan 2024 23:30:39 +0800<![CDATA[C-Eval 大语言模型测评 - lm evaluation harness + vllm 跑测评]]>https://blog.csdn.net/arkohut/article/details/135515727https://blog.csdn.net/arkohut/article/details/135515727arkohutWed, 10 Jan 2024 23:41:03 +0800<![CDATA[Yi-34B 本地部署简单测试]]>https://blog.csdn.net/arkohut/article/details/135331469https://blog.csdn.net/arkohut/article/details/135331469arkohutTue, 02 Jan 2024 01:54:20 +0800<![CDATA[双卡 3080ti 部署 Yi-34B 大模型 - Gradio + vLLM 踩坑全记录]]>https://blog.csdn.net/arkohut/article/details/135321242https://blog.csdn.net/arkohut/article/details/135321242arkohutMon, 01 Jan 2024 17:37:10 +0800<![CDATA[只需 24G 显存，用 vllm 跑起来 Yi-34B 中英双语大模型]]>https://blog.csdn.net/arkohut/article/details/135274973https://blog.csdn.net/arkohut/article/details/135274973arkohutThu, 28 Dec 2023 17:39:08 +0800<![CDATA[用 llama.cpp 跑通 mixtral MoE 模型]]>https://blog.csdn.net/arkohut/article/details/135168655https://blog.csdn.net/arkohut/article/details/135168655arkohutSat, 23 Dec 2023 14:42:52 +0800<![CDATA[VLLM 测试 Mixtral MoE 的 GPTQ 量化版本]]>https://blog.csdn.net/arkohut/article/details/135167762https://blog.csdn.net/arkohut/article/details/135167762arkohutSat, 23 Dec 2023 13:44:22 +0800