http://luispa.com/en/tags/hugging-face/Recent content in Hugging-Face on Technical NotesHugo -- 0.148.0en-USSun, 22 Feb 2026 00:00:00 +0000http://luispa.com/en/posts/2026-02-22-limitar-hf/Sun, 22 Feb 2026 00:00:00 +0000http://luispa.com/en/posts/2026-02-22-limitar-hf/<img src="http://luispa.com/img/posts/logo-shaping.svg" alt="Bandwidth Limiting Logo" width="150px" height="150px" style="float:left; padding-right:25px" /> <p>Downloading LLM models locally is something you do occasionally, but when you pull a massive 122B parameter model like <code>Sehyo/Qwen3.5-122B-A10B-NVFP4</code>, the download hogs the entire connection and leaves the rest of the household without internet. The Hugging Face CLI (<code>huggingface-cli</code> or <code>hf</code>) doesn&rsquo;t have a <code>--limit-rate</code> flag, so you need to find alternatives. In this post I explain two ways to limit bandwidth on Linux using Docker (my preferred method) or Wondershaper at the host level.</p> <br clear="left"/>