diff --git a/.github/workflows/image.yml b/.github/workflows/image.yml
deleted file mode 100644
index 4afa1eb..0000000
--- a/.github/workflows/image.yml
+++ /dev/null
@@ -1,171 +0,0 @@
-name: "Build image and test"
-
-on:
-  schedule:
-    - cron: "20 4 * * 1" # once a week
-  workflow_dispatch:
-
-  push:
-    branches: [main]
-
-  pull_request:
-
-concurrency:
-  group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
-  cancel-in-progress: true
-
-env:
-  KIND_CLUSTER_NAME: "kserve-testing"
-  ISVC_NAME: "vllm-tgis-isvc"
-  TEST_NS: "kserve-demo"
-  # note: knative serving will try to resolve the image tag unless the registry is kind.local
-  # See the deployment configmap:
-  #   registries-skipping-tag-resolving: "kind.local,ko.local,dev.local"
-  DEV_IMAGE: "kind.local/vllm-tgis:dev"
-  QUAY_IMAGE: "quay.io/dtrifiro/vllm-tgis"
-
-jobs:
-  build-image:
-    timeout-minutes: 40
-    name: "Build image"
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v4
-
-      - name: Free Disk Space
-        uses: jlumbroso/free-disk-space@v1.3.1
-        with:
-          tool-cache: false
-          large-packages: false
-          docker-images: false
-
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
-      # TODO: setup caching for mount=type=cache ?
-      # - .nox (nox envs)
-      # - /root/.cache/pip
-      # https://docs.docker.com/build/ci/github-actions/cache/#cache-mounts
-
-      - name: Login to Quay
-        uses: docker/login-action@v3
-        with:
-          registry: quay.io
-          username: ${{ secrets.QUAY_USERNAME }}
-          password: ${{ secrets.QUAY_TOKEN }}
-
-      - name: Build and export
-        uses: docker/build-push-action@v5
-        with:
-          context: .
-          file: ./Dockerfile
-          tags: ${{ env.DEV_IMAGE }}
-          # outputs: type=oci,dest=/tmp/image.tar
-          outputs: type=docker,dest=/tmp/image.tar
-          cache-from: type=gha
-          cache-to: type=gha,mode=max
-
-      - name: Docker meta
-        id: meta
-        uses: docker/metadata-action@v5
-        with:
-          images: ${{ env.QUAY_IMAGE }}
-          tags: |
-            type=raw,value=latest,enable=${{ github.ref == format('refs/heads/{0}', 'main') }}
-            type=sha
-            type=ref,event=branch
-            type=ref,event=pr
-
-      - name: Push image
-        uses: docker/build-push-action@v5
-        with:
-          context: .
-          file: ./Dockerfile
-          load: true
-          cache-from: type=gha
-          tags: ${{ steps.meta.outputs.tags }}
-          labels: ${{ steps.meta.outputs.labels }}
-
-      - name: "Push to quay" # NOTE: we need this step since build and push hangs on push
-        run: |
-          docker push --all-tags "${QUAY_IMAGE}"
-
-      # - name: Upload artifact
-      #   uses: actions/upload-artifact@v4
-      #   with:
-      #     name: vllm-tgis
-      #     path: /tmp/image.tar
-
-  # Temporarily disabled: need to make sure the runner can run vLLM (AVX512 instruction set or GPU)
-  # kserve-smoke-test:
-  #   needs: build-image
-
-  #   name: Kserve Smoke Test
-  #   runs-on: ubuntu-latest
-  #   steps:
-  #     - uses: actions/checkout@v4
-
-  #     - name: Setup kind/kserve
-  #       uses: dtrifiro/setup-kserve@v0.0.2
-  #       with:
-  #         cluster_name: ${{ env.KIND_CLUSTER_NAME }}
-  #         namespace: ${{ env.TEST_NS }}
-
-  #     - name: Download built image
-  #       uses: actions/download-artifact@v4
-
-  #     - name: Load built image into kind
-  #       run: |
-  #         kind load image-archive --name ${{ env.KIND_CLUSTER_NAME }} vllm-tgis/image.tar
-
-  #     - name: Free Disk Space
-  #       uses: jlumbroso/free-disk-space@v1.3.1
-  #       with:
-  #         tool-cache: false
-
-  #     - name: Setup flan-t5-small model volume
-  #       run: |
-  #         sed 's|quay.io/dtrifiro/vllm-tgis:fast|${{ env.DEV_IMAGE }}|g' .github/test/kserve/setup.yaml | \
-  #           kubectl apply -f -
-
-  #         max_retries=10
-  #         wait_time=60s
-  #         until kubectl wait --for=jsonpath='{.status.phase}'=Succeeded pod/setup-flan-t5-small --timeout ${wait_time}; do
-  #           echo "Current status:"
-  #           kubectl describe pod,pv,pvc
-  #           max_retries=$((max_retries-1))
-  #           if [[ $max_retries -le 0 ]]; then
-  #             echo "Failed to setup model"
-  #             kubectl logs pod/setup-flan-t5-small --all-containers
-  #             exit 1
-  #           fi
-  #           echo "-------------------"
-  #         done
-
-  #     - name: Deploy ServingRuntime/InferenceService
-  #       run: |
-  #         sed 's|quay.io/opendatahub/vllm:fast|${{ env.DEV_IMAGE }}|g' .github/test/kserve/vllm-tgis.yaml | \
-  #           kubectl apply -f -
-
-  #         max_retries=10
-  #         wait_time=60s
-  #         until kubectl wait isvc/${ISVC_NAME} --for=condition=Ready --timeout=${wait_time}; do
-  #           echo "Current status:"
-  #           kubectl describe isvc,servingruntime,pod
-  #           max_retries=$((max_retries-1))
-  #           if [[ $max_retries -le 0 ]]; then
-  #             exit 1
-  #           fi
-  #           echo "-------------------"
-  #         done
-
-  #     - name: Perform test inference (http)
-  #       run: |
-  #         export ISVC_URL="$(oc get isvc ${ISVC_NAME} -o jsonpath='{.status.components.predictor.url}')"
-  #         export ISVC_HOSTNAME=$(echo $ISVC_URL | cut -d/ -f3-)
-  #         echo "Querying ISVC at: ${ISVC_URL}"
-
-  #         # We can't query the service via hostname, we need to add the entry to /etc/hosts
-  #         echo "127.0.0.1 ${ISVC_HOSTNAME}" | sudo tee -a /etc/hosts
-
-  #         python examples/inference.py
diff --git a/Dockerfile b/Dockerfile
deleted file mode 100644
index 49ce02e..0000000
--- a/Dockerfile
+++ /dev/null
@@ -1,122 +0,0 @@
-ARG BASE_UBI_IMAGE_TAG=9.3-1612
-ARG PYTHON_VERSION=3.11
-
-FROM registry.access.redhat.com/ubi9/ubi-minimal:${BASE_UBI_IMAGE_TAG} as base
-
-ENV LANG=C.UTF-8 \
-    LC_ALL=C.UTF-8
-
-RUN microdnf install -y \
-    python3.11-pip python3.11-wheel \
-    && microdnf clean all
-
-WORKDIR /workspace
-
-
-FROM base as python-base
-
-ARG PYTHON_VERSION
-
-ENV VIRTUAL_ENV=/opt/vllm
-ENV PATH="$VIRTUAL_ENV/bin:$PATH"
-
-# hadolint ignore=DL3041
-RUN microdnf install -y \
-    git \
-    python${PYTHON_VERSION}-devel python${PYTHON_VERSION}-pip python${PYTHON_VERSION}-wheel && \
-    python${PYTHON_VERSION} -m venv $VIRTUAL_ENV && microdnf clean all
-
-# hadolint ignore=DL3042,DL3013
-RUN --mount=type=cache,target=/root/.cache/pip \
-    pip install -U pip
-
-# TODO: add flash attention build
-
-FROM python-base AS build
-ARG PYTHON_VERSION
-
-ENV VIRTUAL_ENV=/opt/vllm
-ENV PATH="$VIRTUAL_ENV/bin:$PATH"
-
-COPY --from=python-base /opt/vllm /opt/vllm
-
-# hadolint ignore=DL3042
-RUN --mount=type=cache,target=/root/.cache/pip \
-    pip install nox==2023.4.22 # TODO: setup renovate
-
-COPY README.md .
-
-COPY pyproject.toml .
-COPY noxfile.py .
-
-COPY src src
-
-# setuptools scm requires the git directory to infer version from git tags, so we bind-mount the git dir when building
-RUN --mount=type=bind,source=.git,target=.git \
-    --mount=type=cache,target=.nox \
-    --mount=type=cache,target=/root/.cache/pip \
-    nox -s build-${PYTHON_VERSION}
-
-
-FROM base AS deploy
-ARG flash_attn_version=2.5.8
-ARG cuda_version_flashattn=122
-ARG torch_version=2.1
-
-WORKDIR /workspace
-
-COPY --from=python-base /opt/vllm /opt/vllm
-
-ENV VIRTUAL_ENV=/opt/vllm
-ENV PATH=$VIRTUAL_ENV/bin/:$PATH
-
-# Triton needs a CC compiler
-# hadolint ignore=DL3041
-RUN microdnf install -y gcc \
-    && microdnf clean all
-
-ENV FLASH_ATTN_VERSION=${flash_attn_version}
-ENV CUDA_VERSION_FLASHATTN=${cuda_version_flashattn}
-ENV CUDA_VERSION="12.0.0"
-ENV TORCH_VERSION=${torch_version}
-
-
-# make sure that the python version in the flash-attn wheel install below matches ${PYTHON_VERSION}
-# hadolint ignore=DL3042
-RUN --mount=type=cache,target=/root/.cache/pip \
-    --mount=type=bind,from=build,src=/workspace/dist/,target=/workspace/dist/ \
-    pip install \
-        https://github.com/Dao-AILab/flash-attention/releases/download/v${FLASH_ATTN_VERSION}/flash_attn-${FLASH_ATTN_VERSION}+cu${CUDA_VERSION_FLASHATTN}torch${TORCH_VERSION}cxx11abiFALSE-cp311-cp311-linux_x86_64.whl \
-        dist/*whl
-
-# vllm requires a specific nccl version built from source distribution
-# See https://github.com/NVIDIA/nccl/issues/1234
-RUN pip install \
-        -v \
-        --force-reinstall \
-        --no-binary="all" \
-        --no-cache-dir \
-        "vllm-nccl-cu12==2.18.1.0.4.0" && \
-    mv /root/.config/vllm/nccl/cu12/libnccl.so.2.18.1 /opt/vllm/lib/ && \
-    chmod 0755 /opt/vllm/lib/libnccl.so.2.18.1
-
-ENV HF_HUB_OFFLINE=1 \
-    PORT=8000 \
-    GRPC_PORT=8033 \
-    HOME=/home/vllm \
-    VLLM_NCCL_SO_PATH=/opt/vllm/lib/libnccl.so.2.18.1 \
-    VLLM_USAGE_SOURCE=production-docker-image \
-    VLLM_WORKER_MULTIPROC_METHOD=fork
-
-# setup non-root user for OpenShift
-RUN microdnf install -y shadow-utils \
-    && umask 002 \
-    && useradd --uid 2000 --gid 0 vllm \
-    && microdnf remove -y shadow-utils \
-    && microdnf clean all \
-    && chmod g+rwx $HOME /usr/src /workspace
-
-COPY LICENSE /licenses/vllm.md
-
-USER 2000
-CMD ["python", "-m", "vllm_tgis_adapter"]
diff --git a/README.md b/README.md
index e7694ec..42c33fa 100644
--- a/README.md
+++ b/README.md
@@ -3,9 +3,8 @@
 vLLM adapter for a TGIS-compatible grpc server.
 
 [![PyPi](https://img.shields.io/pypi/v/vllm-tgis-adapter?label=pip)](https://pypi.org/project/vllm-tgis-adapter)
-[![Tests](https://github.com/dtrifiro/vllm-tgis-adapter/actions/workflows/tests.yaml/badge.svg)](https://github.com/dtrifiro/vllm-tgis-adapter/actions/workflows/tests.yaml)
-[![Docker Image Build](https://github.com/dtrifiro/vllm-tgis-adapter/actions/workflows/image.yml/badge.svg)](https://github.com/dtrifiro/vllm-tgis-adapter/actions/workflows/image.yml)
-[![quay.io/dtrifiro/vllm-tgis](https://img.shields.io/badge/quay.io-dtrifiro/vllm--tgis-darkred)](https://quay.io/repository/dtrifiro/vllm-tgis?tab=tags)
+[![Tests](https://github.com/opendatahub-io/vllm-tgis-adapter/actions/workflows/tests.yaml/badge.svg)](https://github.com/opendatahub-io/vllm-tgis-adapter/actions/workflows/tests.yaml)
+[![quay.io/opendatahub/vllm](https://img.shields.io/badge/quay.io-opendatahub/vllm--tgis-darkred)](https://quay.io/repository/opendatahub/vllm?tab=tags)
 
 ## Install
 
@@ -49,10 +48,10 @@ bash examples/inference.sh
 
 ### Docker
 
-Image available at [quay.io/dtrifiro/vllm-tgis](https://quay.io/dtrifiro/vllm-tgis?tab=tags)
+Image available at [quay.io/opendatahub/vllm](https://quay.io/opendatahub/vllm?tab=tags), built from [opendatahub-io/vllm](https://github.com/opendatahub-io/vllm)'s [Dockerfile.ubi](https://github.com/opendatahub-io/vllm/tree/main/Dockerfile.ubi)
 
 ```bash
-docker pull quay.io/dtrifiro/vllm-tgis
+docker pull quay.io/opendatahub/vllm
 ```
 
 ### Inference
diff --git a/pyproject.toml b/pyproject.toml
index fd15591..23aafa6 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -37,8 +37,8 @@ dependencies = [
 ]
 
 [project.urls]
-Issues = "https://github.com/dtrifiro/vllm_tgis_adapter/issues"
-Source = "https://github.com/dtrifiro/vllm_tgis_adapter"
+Issues = "https://github.com/opendatahub-io/vllm_tgis_adapter/issues"
+Source = "https://github.com/opendatahub-io/vllm_tgis_adapter"
 
 [project.scripts]
 grpc_healthcheck = "vllm_tgis_adapter.healthcheck:cli"