From f491096f7d55a72b2ac364ca14668d6f577ad8fc Mon Sep 17 00:00:00 2001
From: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com>
Date: Thu, 28 Nov 2024 16:31:36 +0100
Subject: [PATCH] Fix docker CI : install autogptq from source (#35000)

* Fixed Docker

* Test ci

* Finally

* add comment
---
 docker/transformers-quantization-latest-gpu/Dockerfile | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/docker/transformers-quantization-latest-gpu/Dockerfile b/docker/transformers-quantization-latest-gpu/Dockerfile
index a8f131164eb4ae..089be4a4460101 100755
--- a/docker/transformers-quantization-latest-gpu/Dockerfile
+++ b/docker/transformers-quantization-latest-gpu/Dockerfile
@@ -36,8 +36,10 @@ RUN python3 -m pip install --no-cache-dir einops
 # Add bitsandbytes for mixed int8 testing
 RUN python3 -m pip install --no-cache-dir bitsandbytes
 
-# Add auto-gptq for gtpq quantization testing
-RUN python3 -m pip install --no-cache-dir auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/
+# Add auto-gptq for gtpq quantization testing, installed from source for pytorch==2.5.1 compatibility
+# TORCH_CUDA_ARCH_LIST="7.5+PTX" is added to make the package compile for Tesla T4 gpus available for the CI.
+RUN pip install gekko
+RUN git clone https://github.com/PanQiWei/AutoGPTQ.git && cd AutoGPTQ && TORCH_CUDA_ARCH_LIST="7.5+PTX" python3 setup.py install
 
 # Add optimum for gptq quantization testing
 RUN python3 -m pip install --no-cache-dir git+https://github.com/huggingface/optimum@main#egg=optimum