diff --git a/packages/text-embeddings/Dockerfile b/packages/text-embeddings/Dockerfile
index 9c1e76f5b0..9af7c55376 100644
--- a/packages/text-embeddings/Dockerfile
+++ b/packages/text-embeddings/Dockerfile
@@ -14,7 +14,6 @@ ENV PATH="/leapfrogai/.venv/bin:$PATH"
 # copy and install all python dependencies
 # NOTE: We are copying the leapfrog whl to this filename because installing 'optional extras' from
 #       a wheel requires the absolute path to the wheel file (instead of a wildcard whl)
-
 COPY --from=sdk /leapfrogai/${SDK_DEST} ${SDK_DEST}
 COPY packages/text-embeddings packages/text-embeddings
 
@@ -22,14 +21,6 @@ RUN rm -f packages/text-embeddings/build/*.whl
 RUN python -m pip wheel packages/text-embeddings -w packages/text-embeddings/build --find-links=${SDK_DEST}
 RUN pip install packages/text-embeddings/build/lfai_text_embeddings*.whl --no-index --find-links=packages/text-embeddings/build/
 
-
-# download model
-RUN python -m pip install -U huggingface_hub[cli,hf_transfer]
-ARG REPO_ID="hkunlp/instructor-xl"
-ARG REVISION="ce48b213095e647a6c3536364b9fa00daf57f436"
-COPY packages/text-embeddings/scripts/model_download.py scripts/model_download.py
-RUN REPO_ID=${REPO_ID} REVISION=${REVISION} python scripts/model_download.py
-
 # hardened and slim python image
 FROM ghcr.io/defenseunicorns/leapfrogai/python:3.11
 
@@ -38,7 +29,6 @@ ENV PATH="/leapfrogai/.venv/bin:$PATH"
 WORKDIR /leapfrogai
 
 COPY --from=builder /leapfrogai/.venv/ /leapfrogai/.venv/
-COPY --from=builder /leapfrogai/.model/ /leapfrogai/.model/
 
 COPY packages/text-embeddings/main.py .
 
diff --git a/packages/text-embeddings/chart/templates/deployment.yaml b/packages/text-embeddings/chart/templates/deployment.yaml
index 8d014f1c72..13869b0a5f 100644
--- a/packages/text-embeddings/chart/templates/deployment.yaml
+++ b/packages/text-embeddings/chart/templates/deployment.yaml
@@ -23,8 +23,38 @@ spec:
         {{- toYaml . | nindent 8 }}
       {{- end }}
       labels:
+        app: lfai-text-embeddings
         {{- include "chart.selectorLabels" . | nindent 8 }}
     spec:
+      # It's necessary to include the ###ZARF_DATA_INJECTION_MARKER### somewhere in the podspec, otherwise data injections will not occur.
+      initContainers:
+        - name: data-loader
+          image: cgr.dev/chainguard/bash:latest
+          securityContext:
+            runAsUser: 65532
+            runAsGroup: 65532
+            fsGroup: 65532
+          # This command looks for the Zarf "data injection marker" which is a timestamped file that is injected after everything else and marks the injection as complete.
+          command:
+            [
+              "sh",
+              "-c",
+              'while [ ! -f /data/.model/###ZARF_DATA_INJECTION_MARKER### ]; do echo "waiting for zarf data sync" && sleep 1; done; echo "we are done waiting!"',
+            ]
+          resources:
+            requests:
+              memory: "64Mi"
+              cpu: "200m"
+            limits:
+              memory: "128Mi"
+              cpu: "500m"
+          volumeMounts:
+            - name: leapfrogai-pv-storage
+              mountPath: /data
+      volumes:
+        - name: leapfrogai-pv-storage
+          persistentVolumeClaim:
+            claimName: lfai-text-embeddings-pv-claim
       securityContext:
         {{- toYaml .Values.podSecurityContext | nindent 8 }}
       containers:
@@ -39,6 +69,12 @@ spec:
               protocol: TCP
           resources:
             {{- toYaml .Values.resources | nindent 12 }}
+          volumeMounts:
+            - name: leapfrogai-pv-storage
+              mountPath: "/data"
+          env:
+            - name: LFAI_MODEL_DIR
+              value: '/data/.model'
       {{- with .Values.nodeSelector }}
       nodeSelector:
         {{- toYaml . | nindent 8 }}
diff --git a/packages/text-embeddings/chart/templates/pvc.yaml b/packages/text-embeddings/chart/templates/pvc.yaml
new file mode 100644
index 0000000000..fa7e02b675
--- /dev/null
+++ b/packages/text-embeddings/chart/templates/pvc.yaml
@@ -0,0 +1,12 @@
+apiVersion: v1
+kind: PersistentVolumeClaim
+metadata:
+    name: lfai-text-embeddings-pv-claim
+    namespace: leapfrogai
+spec:
+  storageClassName: "local-path"
+  accessModes:
+    - ReadWriteOnce
+  resources:
+    requests:
+      storage: 15Gi
diff --git a/packages/text-embeddings/main.py b/packages/text-embeddings/main.py
index 6af2d6a720..0ad8ce8242 100644
--- a/packages/text-embeddings/main.py
+++ b/packages/text-embeddings/main.py
@@ -1,5 +1,6 @@
 import asyncio
 import logging
+import os
 
 from InstructorEmbedding import INSTRUCTOR
 from leapfrogai_sdk import (
@@ -10,7 +11,8 @@
     serve,
 )
 
-model = INSTRUCTOR("./.model")
+model_dir = os.environ.get("LFAI_MODEL_PATH", ".model")
+model = INSTRUCTOR(model_dir)
 
 
 class InstructorEmbedding:
diff --git a/packages/text-embeddings/zarf.yaml b/packages/text-embeddings/zarf.yaml
index d3bec755cf..b47672c31e 100644
--- a/packages/text-embeddings/zarf.yaml
+++ b/packages/text-embeddings/zarf.yaml
@@ -32,3 +32,20 @@ components:
           - "embedding-values.yaml"
     images:
       - ghcr.io/defenseunicorns/leapfrogai/text-embeddings:###ZARF_PKG_TMPL_IMAGE_VERSION###
+      - cgr.dev/chainguard/bash:latest
+    dataInjections:
+      - source: .model/
+        target:
+          namespace: leapfrogai
+          selector: app=lfai-text-embeddings
+          container: data-loader
+          path: /data/.model
+        compress: true
+    actions:
+      onCreate:
+        before:
+          # NOTE: This assumes python is installed and in $PATH
+          - cmd: python scripts/model_download.py
+            env:
+              - REPO_ID=hkunlp/instructor-xl
+              - REVISION=ce48b213095e647a6c3536364b9fa00daf57f436