From 1b34e4037375d9d7a52aa9a9ab95bd77dbaa34da Mon Sep 17 00:00:00 2001 From: Nikhil Gupta Date: Tue, 11 Nov 2025 15:54:37 +0000 Subject: [PATCH] [fix]: fix env issues with vLLM int4 acceleration LP Signed-off-by: Nikhil Gupta --- .../vllm-acceleration/1-overview-and-build.md | 7 ++++--- .../vllm-acceleration/2-quantize-model.md | 2 +- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/content/learning-paths/servers-and-cloud-computing/vllm-acceleration/1-overview-and-build.md b/content/learning-paths/servers-and-cloud-computing/vllm-acceleration/1-overview-and-build.md index bc4bfb3e3..9209cebfc 100644 --- a/content/learning-paths/servers-and-cloud-computing/vllm-acceleration/1-overview-and-build.md +++ b/content/learning-paths/servers-and-cloud-computing/vllm-acceleration/1-overview-and-build.md @@ -42,7 +42,8 @@ Install the minimum system package used by vLLM on Arm: ```bash sudo apt-get update -y -sudo apt-get install -y libnuma-dev +sudo apt-get install -y build-essential cmake libnuma-dev +sudo apt install python3.12-venv python3.12-dev ``` Optional performance helper you can install now or later: @@ -60,9 +61,9 @@ On aarch64, vLLM’s CPU backend automatically builds with Arm Compute Library v Create and activate a virtual environment: ```bash -python3 -m venv vllm_env +python3.12 -m venv vllm_env source vllm_env/bin/activate -python -m pip install --upgrade pip +python3 -m pip install --upgrade pip ``` Clone vLLM and install build requirements: diff --git a/content/learning-paths/servers-and-cloud-computing/vllm-acceleration/2-quantize-model.md b/content/learning-paths/servers-and-cloud-computing/vllm-acceleration/2-quantize-model.md index a5d472ccc..056010811 100644 --- a/content/learning-paths/servers-and-cloud-computing/vllm-acceleration/2-quantize-model.md +++ b/content/learning-paths/servers-and-cloud-computing/vllm-acceleration/2-quantize-model.md @@ -135,7 +135,7 @@ This script creates a Arm KleidiAI 4‑bit quantized copy of the vLLM model and ```bash # DeepSeek example -python quantize_vllm_models.py deepseek-ai/DeepSeek-V2-Lite \ +python3 quantize_vllm_models.py deepseek-ai/DeepSeek-V2-Lite \ --scheme channelwise --method mse ```