diff --git a/distributions/meta-reference-gpu/build.yaml b/distributions/meta-reference-gpu/build.yaml
index e76197330..5b1521a92 100644
--- a/distributions/meta-reference-gpu/build.yaml
+++ b/distributions/meta-reference-gpu/build.yaml
@@ -1,5 +1,6 @@
 name: meta-reference-gpu
 distribution_spec:
+  docker_image: pytorch/pytorch
   description: Use code from `llama_stack` itself to serve all llama stack APIs
   providers:
     inference: meta-reference