Update README.md

README.md

@@ -61,27 +61,7 @@ The model was evaluated on GSM8K benchmarks.
 
 ### Reproduction
 
-The GSM8K results were obtained using the `lm-evaluation-harness` framework, based on the Docker image `rocm/vllm-dev:base`, with vLLM and lm-eval compiled and installed from source inside the container.
-
-#### Preparation in container
-```
-# installl vLLM
-git clone https://github.com/vllm-project/vllm.git
-cd vllm
-git checkout 74c583bc508c2dafb9e95bab3b635884e4a021f3
-pip install -r requirements/rocm.txt
-python setup.py develop
-cd ..
-
-# install lm-eval
-git clone --recursive https://github.com/EleutherAI/lm-evaluation-harness.git
-cd lm-evaluation-harness
-git checkout 3ac28fc5782dc1b3cc62ae4337b1aaf8d0065bef
-pip install -e .
-cd ..
-
-pip install lm-eval[api]
-```
+The GSM8K results were obtained using the lm-evaluation-harness framework, based on the Docker image `rocm/vllm-private:vllm_dev_base_mxfp4_20260122`, with vLLM, lm-eval and amd-quark compiled and installed from source inside the image.
 
 #### Launching server
 ```