Error: Enable code enrichment Enable formula enrichment

Hello, I use this configuration. Everything is installed and working.

When I turn on Enable code enrichment Enable formula enrichment an error occurs:

NFO: 127.0.0.1:40232 - "GET /v1alpha/status/poll/23553a3d-a6a6-44a7-9ecd-8729765448fe?wait=5 HTTP/1.1" 200 OK
ERROR:docling_serve.engines.async_local.worker:Worker 0 failed to process job 23553a3d-a6a6-44a7-9ecd-8729765448fe: 500: Repo id must be in the form 'repo_name' or 'namespace/repo_name': '/opt/app-root/src/.cache/docling/models/ds4sd--CodeFormula'. Use repo_type argument if needed.
INFO: 127.0.0.1:40248 - "GET /v1alpha/status/poll/23553a3d-a6a6-44a7-9ecd-8729765448fe?wait=5 HTTP/1.1" 200 OK
ERROR:docling_serve.gradio_ui:Error processing file(s): Task failed with status 'failure'
INFO: 127.0.0.1:53328 - "GET /ui/static/fonts/ui-sans-serif/ui-sans-serif-Bold.woff2 HTTP/1.1" 404 Not Found
INFO: 127.0.0.1:53328 - "GET /ui/static/fonts/system-ui/system-ui-Bold.woff2 HTTP/1.1" 404 Not Found.

Here is the configuration file:

apiVersion: v1
kind: Namespace
metadata:
name: docling
---
apiVersion: v1
kind: PersistentVolume
metadata:
name: docling-model-cache-pv
namespace: docling
spec:
capacity:
storage: 10Gi
accessModes:
- ReadWriteOnce
persistentVolumeReclaimPolicy: Retain
storageClassName: microk8s-hostpath
hostPath:
path: /mnt/disks/modelcache
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: docling-model-cache-pvc
namespace: docling
spec:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 10Gi
storageClassName: microk8s-hostpath
---
apiVersion: batch/v1
kind: Job
metadata:
name: docling-model-cache-load
namespace: docling
spec:
selector: {}
template:
metadata:
name: docling-model-load
spec:
securityContext:
fsGroup: 1000
restartPolicy: Never
initContainers:
- name: init-permissions
image: busybox
command:
- sh
- -c
- |
mkdir -p /modelcache/.cache && chmod -R 777 /modelcache
volumeMounts:
- name: modelcache
mountPath: /modelcache
containers:
- name: loader
image: ghcr.io/docling-project/docling-serve-cpu:main
command:
- docling-tools
- models
- download
- --output-dir=/modelcache
- layout
- tableformer
- code_formula
- picture_classifier
- smolvlm
- granite_vision
- easyocr
env:
- name: HF_HOME
value: "/modelcache/hf_home"
- name: HUGGINGFACE_HUB_CACHE
value: "/modelcache/hf_home/hub"
volumeMounts:
- name: modelcache
mountPath: /modelcache
volumes:
- name: modelcache
persistentVolumeClaim:
claimName: docling-model-cache-pvc
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: docling-serve
namespace: docling
labels:
app: docling-serve
spec:
replicas: 1
selector:
matchLabels:
app: docling-serve
template:
metadata:
labels:
app: docling-serve
spec:
securityContext:
fsGroup: 1000
restartPolicy: Always
initContainers:
- name: init-permissions
image: busybox
command:
- sh
- -c
- |
mkdir -p /modelcache/.cache && chmod -R 777 /modelcache
volumeMounts:
- name: modelcache
mountPath: /modelcache
containers:
- name: api
image: ghcr.io/docling-project/docling-serve-cpu:main
imagePullPolicy: Always
ports:
- containerPort: 5001
env:
- name: DOCLING_SERVE_ENABLE_UI
value: "true"
- name: DOCLING_ARTIFACTS_PATH
value: "/modelcache"
- name: HF_HOME
value: "/modelcache/hf_home"
- name: HUGGINGFACE_HUB_CACHE
value: "/modelcache/hf_home/hub"
resources:
requests:
cpu: 250m
memory: 1Gi
limits:
cpu: 500m
memory: 2Gi
volumeMounts:
- name: modelcache
mountPath: /modelcache
volumes:
- name: modelcache
persistentVolumeClaim:
claimName: docling-model-cache-pvc

How can I solve the problem? Thank you.

Start with the Kubernetes manifest’s model-cache Job and Deployment, especially the `docling-tools models download` entry point and `DOCLING_ARTIFACTS_PATH`; reproduce with code and formula enrichment enabled and inspect the cached model path shown in the worker error. Done means the worker processes a file successfully without the repository-ID failure.

Written by the indexing model from the issue text.