rapidsai · rapids-bot · Jan 26, 2026 · Jan 26, 2026
@@ -22,9 +22,6 @@ on:
       CUVS_BENCH_TAG:
         required: true
         type: string
-      CUVS_BENCH_DATASETS_TAG:
-        required: true
-        type: string
       CUVS_BENCH_CPU_TAG:
         required: true
         type: string
@@ -110,18 +107,6 @@ jobs:
           tags: ${{ inputs.CUVS_BENCH_TAG }}-${{ matrix.ARCH }}
           # ensure only OCI mediatypes are used: https://docs.docker.com/build/exporters/#oci-media-types
           outputs: type=registry,oci-mediatypes=true
-      # - name: Build cuVS Benchmarks GPU with datasets image
-      #   uses: docker/build-push-action@v6
-      #   with:
-      #     context: context
-      #     file: cuvs-bench/gpu/Dockerfile
-      #     target: cuvs-bench-datasets
-      #     push: true
-      #     pull: true
-      #     build-args: |
-      #       ${{ steps.generate-build-args.outputs.DOCKER_BUILD_ARGS }}
-      #     tags: ${{ inputs.CUVS_BENCH_DATASETS_TAG }}-${{ matrix.ARCH }}
-      #     outputs: type=registry,oci-mediatypes=true
       - name: Build cuVS Benchmarks CPU image
         if: inputs.BUILD_CUVS_BENCH_CPU_IMAGE
         uses: docker/build-push-action@v6

@@ -59,14 +59,12 @@ jobs:
       BASE_IMAGE_REPO: ${{ steps.compute-image-repo.outputs.BASE_IMAGE_REPO }}
       NOTEBOOKS_IMAGE_REPO: ${{ steps.compute-image-repo.outputs.NOTEBOOKS_IMAGE_REPO }}
       CUVS_BENCH_IMAGE_REPO: ${{ steps.compute-image-repo.outputs.CUVS_BENCH_IMAGE_REPO }}
-      CUVS_BENCH_DATASETS_IMAGE_REPO: ${{ steps.compute-image-repo.outputs.CUVS_BENCH_DATASETS_IMAGE_REPO }}
       CUVS_BENCH_CPU_IMAGE_REPO: ${{ steps.compute-image-repo.outputs.CUVS_BENCH_CPU_IMAGE_REPO }}
       MATRIX: ${{ steps.compute-matrix.outputs.MATRIX }}
       TEST_MATRIX: ${{ steps.compute-test-matrix.outputs.TEST_MATRIX }}
       BASE_TAG_PREFIX: ${{ steps.compute-tag-prefix.outputs.BASE_TAG_PREFIX }}
       NOTEBOOKS_TAG_PREFIX: ${{ steps.compute-tag-prefix.outputs.NOTEBOOKS_TAG_PREFIX }}
       CUVS_BENCH_TAG_PREFIX: ${{ steps.compute-tag-prefix.outputs.CUVS_BENCH_TAG_PREFIX }}
-      CUVS_BENCH_DATASETS_TAG_PREFIX: ${{ steps.compute-tag-prefix.outputs.CUVS_BENCH_DATASETS_TAG_PREFIX }}
       CUVS_BENCH_CPU_TAG_PREFIX: ${{ steps.compute-tag-prefix.outputs.CUVS_BENCH_CPU_TAG_PREFIX }}
       RAPIDS_VER: ${{ steps.compute-rapids-ver.outputs.RAPIDS_VER }}
       ALPHA_TAG: ${{ steps.compute-rapids-ver.outputs.ALPHA_TAG }}
@@ -89,33 +87,28 @@ jobs:
             BASE_TAG_PREFIX="docker-${pr_num}-"
             NOTEBOOKS_TAG_PREFIX="docker-notebooks-${pr_num}-"
             CUVS_BENCH_TAG_PREFIX="docker-cuvs-bench-${pr_num}-"
-            CUVS_BENCH_DATASETS_TAG_PREFIX="docker-cuvs-bench-datasets-${pr_num}-"
             CUVS_BENCH_CPU_TAG_PREFIX="docker-cuvs-bench-cpu-${pr_num}-"
           fi
           echo "BASE_TAG_PREFIX=${BASE_TAG_PREFIX}" | tee -a ${GITHUB_OUTPUT}
           echo "NOTEBOOKS_TAG_PREFIX=${NOTEBOOKS_TAG_PREFIX}" | tee -a ${GITHUB_OUTPUT}
           echo "CUVS_BENCH_TAG_PREFIX=${CUVS_BENCH_TAG_PREFIX}" | tee -a ${GITHUB_OUTPUT}
-          echo "CUVS_BENCH_DATASETS_TAG_PREFIX=${CUVS_BENCH_DATASETS_TAG_PREFIX}" | tee -a ${GITHUB_OUTPUT}
           echo "CUVS_BENCH_CPU_TAG_PREFIX=${CUVS_BENCH_CPU_TAG_PREFIX}" | tee -a ${GITHUB_OUTPUT}
       - name: Compute image repo
         id: compute-image-repo
         run: |
           base_repo="base"
           notebooks_repo="notebooks"
           cuvs_bench_repo="cuvs-bench"
-          cuvs_bench_datasets_repo="cuvs-bench-datasets"
           cuvs_bench_cpu_repo="cuvs-bench-cpu"
           if [ "${{ inputs.build_type }}" = "pull-request" ]; then
             base_repo="staging"
             notebooks_repo="staging"
             cuvs_bench_repo="staging"
-            cuvs_bench_datasets_repo="staging"
             cuvs_bench_cpu_repo="staging"
           fi
           echo "BASE_IMAGE_REPO=${base_repo}" | tee -a ${GITHUB_OUTPUT}
           echo "NOTEBOOKS_IMAGE_REPO=${notebooks_repo}" | tee -a ${GITHUB_OUTPUT}
           echo "CUVS_BENCH_IMAGE_REPO=${cuvs_bench_repo}" | tee -a ${GITHUB_OUTPUT}
-          echo "CUVS_BENCH_DATASETS_IMAGE_REPO=${cuvs_bench_datasets_repo}" | tee -a ${GITHUB_OUTPUT}
           echo "CUVS_BENCH_CPU_IMAGE_REPO=${cuvs_bench_cpu_repo}" | tee -a ${GITHUB_OUTPUT}
       - name: Compute RAPIDS_VER
         id: compute-rapids-ver
@@ -198,13 +191,6 @@ jobs:
         ${{ needs.compute-matrix.outputs.ALPHA_TAG }}-\
         cuda${{ matrix.CUDA_TAG }}-\
         py${{ matrix.PYTHON_VER }}"
-      CUVS_BENCH_DATASETS_TAG:
-        "rapidsai/${{ needs.compute-matrix.outputs.CUVS_BENCH_DATASETS_IMAGE_REPO }}:\
-        ${{ needs.compute-matrix.outputs.CUVS_BENCH_DATASETS_TAG_PREFIX }}\
-        ${{ needs.compute-matrix.outputs.RAPIDS_VER }}\
-        ${{ needs.compute-matrix.outputs.ALPHA_TAG }}-\
-        cuda${{ matrix.CUDA_TAG }}-\
-        py${{ matrix.PYTHON_VER }}"
       CUVS_BENCH_CPU_TAG:
         "rapidsai/${{ needs.compute-matrix.outputs.CUVS_BENCH_CPU_IMAGE_REPO }}:\
         ${{ needs.compute-matrix.outputs.CUVS_BENCH_CPU_TAG_PREFIX }}\
@@ -270,8 +256,6 @@ jobs:
           CUVS_BENCH_CPU_IMAGE_BUILT: ${{ matrix.BUILD_CUVS_BENCH_CPU_IMAGE }}
           CUVS_BENCH_IMAGE_REPO: ${{ needs.compute-matrix.outputs.CUVS_BENCH_IMAGE_REPO }}
           CUVS_BENCH_TAG_PREFIX: ${{ needs.compute-matrix.outputs.CUVS_BENCH_TAG_PREFIX }}
-          CUVS_BENCH_DATASETS_IMAGE_REPO: ${{ needs.compute-matrix.outputs.CUVS_BENCH_DATASETS_IMAGE_REPO }}
-          CUVS_BENCH_DATASETS_TAG_PREFIX: ${{ needs.compute-matrix.outputs.CUVS_BENCH_DATASETS_TAG_PREFIX }}
           CUVS_BENCH_CPU_IMAGE_REPO: ${{ needs.compute-matrix.outputs.CUVS_BENCH_CPU_IMAGE_REPO }}
           CUVS_BENCH_CPU_TAG_PREFIX: ${{ needs.compute-matrix.outputs.CUVS_BENCH_CPU_TAG_PREFIX }}
           GPUCIBOT_DOCKERHUB_USER: ${{ secrets.GPUCIBOT_DOCKERHUB_USER }}

@@ -1,5 +1,5 @@
 #!/bin/bash
-# Copyright (c) 2025, NVIDIA CORPORATION.
+# Copyright (c) 2025-2026, NVIDIA CORPORATION.
 
 set -eEuo pipefail
 
@@ -8,26 +8,20 @@ common_path="$(dirname "$(realpath "$0")")/common.sh"
 source "$common_path"
 
 cuvs_bench_source_tags=()
-# cuvs_bench_datasets_source_tags=()
 cuvs_bench_cpu_source_tags=()
 
 # Define tag arrays for different images
 cuvs_bench_tag="${CUVS_BENCH_TAG_PREFIX}${RAPIDS_VER}${ALPHA_TAG}-cuda${CUDA_TAG}-py${PYTHON_VER}"
-# cuvs_bench_datasets_tag="${CUVS_BENCH_DATASETS_TAG_PREFIX}${RAPIDS_VER}${ALPHA_TAG}-cuda${CUDA_TAG}-py${PYTHON_VER}"
 cuvs_bench_cpu_tag="${CUVS_BENCH_CPU_TAG_PREFIX}${RAPIDS_VER}${ALPHA_TAG}-py${PYTHON_VER}"
 
 # Check if all source tags exist and add to source tags array
 for arch in $(echo "${ARCHES}" | jq .[] -r); do
     full_cuvs_bench_tag="${cuvs_bench_tag}-${arch}"
-    # full_cuvs_bench_datasets_tag="${cuvs_bench_datasets_tag}-${arch}"
     full_cuvs_bench_cpu_tag="${cuvs_bench_cpu_tag}-${arch}"
 
     check_tag_exists "$CUVS_BENCH_IMAGE_REPO" "$full_cuvs_bench_tag"
     cuvs_bench_source_tags+=("${org}/${CUVS_BENCH_IMAGE_REPO}:$full_cuvs_bench_tag")
 
-    # check_tag_exists "$CUVS_BENCH_DATASETS_IMAGE_REPO" "$full_cuvs_bench_datasets_tag"
-    # cuvs_bench_datasets_source_tags+=("${org}/${CUVS_BENCH_DATASETS_IMAGE_REPO}:$full_cuvs_bench_datasets_tag")
-
     if [ "$CUVS_BENCH_CPU_IMAGE_BUILT" = "true" ]; then
         check_tag_exists "$CUVS_BENCH_CPU_IMAGE_REPO" "$full_cuvs_bench_cpu_tag"
         cuvs_bench_cpu_source_tags+=("${org}/${CUVS_BENCH_CPU_IMAGE_REPO}:$full_cuvs_bench_cpu_tag")
@@ -37,11 +31,6 @@ done
 docker manifest create "${org}/${CUVS_BENCH_IMAGE_REPO}:${cuvs_bench_tag}" "${cuvs_bench_source_tags[@]}"
 docker manifest push "${org}/${CUVS_BENCH_IMAGE_REPO}:${cuvs_bench_tag}"
 
-# this and everything above that it uses can be uncommented once the issues with cuVS datasets are fixed
-# ref: https://github.com/rapidsai/docker/issues/724
-# docker manifest create "${org}/${CUVS_BENCH_DATASETS_IMAGE_REPO}:${cuvs_bench_datasets_tag}" "${cuvs_bench_datasets_source_tags[@]}"
-# docker manifest push "${org}/${CUVS_BENCH_DATASETS_IMAGE_REPO}:${cuvs_bench_datasets_tag}"
-
 if [ "$CUVS_BENCH_CPU_IMAGE_BUILT" = "true" ]; then
     docker manifest create "${org}/${CUVS_BENCH_CPU_IMAGE_REPO}:${cuvs_bench_cpu_tag}" "${cuvs_bench_cpu_source_tags[@]}"
     docker manifest push "${org}/${CUVS_BENCH_CPU_IMAGE_REPO}:${cuvs_bench_cpu_tag}"

@@ -18,7 +18,6 @@ For complete details, refer to the cuVS documentation: https://docs.rapids.ai/ap
 We provide images for GPU enabled systems, as well as systems without a GPU. The following images are available:
 
 - `cuvs-bench`: Contains GPU and CPU benchmarks, can run all algorithms supported. Will download million-scale datasets as required. Best suited for users that prefer a smaller container size for GPU based systems. Requires the NVIDIA Container Toolkit to run GPU algorithms, can run CPU algorithms without it.
-- `cuvs-bench-datasets`: Contains the GPU and CPU benchmarks with million-scale datasets already included in the container. Best suited for users that want to run multiple million scale datasets already included in the image.
 - `cuvs-bench-cpu`: Contains only CPU benchmarks with minimal size. Best suited for users that want the smallest containers to reproduce benchmarks on systems without a GPU.
 
 Nightly images are located in [DockerHub](https://hub.docker.com/r/rapidsai/cuvs-bench), release versions will be located in NCG in the next release.
@@ -47,7 +46,7 @@ Where:
 
 - `DATA_FOLDER=path/to/store/results/and/data`: Results and datasets will be written to this host folder.
 - `-u $(id -u)`: This flag allows the container to use the host user for permissions
-- `rapidsai/cuvs-bench:26.04a-cuda13-py3.13`: Image to use, either `cuvs-bench` or `cuvs-bench-datasets`, cuVS version, CUDA version, and Python version.
+- `rapidsai/cuvs-bench:26.04a-cuda13-py3.13`: Image to use, `cuvs-bench`, cuVS version, CUDA version, and Python version.
 - "--dataset deep-image-96-angular": Dataset name(s). See https://docs.rapids.ai/api/cuvs/nightly/cuvs_bench for more details.
 - "--normalize": Whether to normalize the dataset, leave string empty ("") to not normalize.
 - "--algorithms cuvs_cagra": What algorithm(s) to use as a ; separated list, as well as any other argument to pass to `cuvs_bench.run`.

@@ -67,17 +67,3 @@ COPY cuvs-bench/run_benchmark.sh /data/scripts/run_benchmark.sh
 CMD ["--dataset fashion-mnist-784-euclidean", "", "--algorithms hnswlib"]
 
 ENTRYPOINT ["/bin/bash", "/data/scripts/run_benchmark.sh"]
-
-FROM cuvs-bench-cpu AS cuvs-bench-cpu-datasets
-
-SHELL ["/bin/bash", "-euo", "pipefail", "-c"]
-
-COPY cuvs-bench/get_datasets.sh /home/rapids/cuvs-bench/get_datasets.sh
-
-COPY cuvs-bench/run_benchmark.sh /data/scripts/run_benchmark_preloaded_datasets.sh
-
-RUN /home/rapids/cuvs-bench/get_datasets.sh
-
-CMD ["--dataset fashion-mnist-784-euclidean", "", "--algorithms hnswlib"]
-
-ENTRYPOINT ["/bin/bash", "/data/scripts/run_benchmark_preloaded_datasets.sh"]
@@ -113,17 +113,3 @@ COPY cuvs-bench/run_benchmark.sh /data/scripts/run_benchmark.sh
 CMD ["--dataset fashion-mnist-784-euclidean", "", "--algorithms cuvs_cagra", ""]
 
 ENTRYPOINT ["/bin/bash", "/data/scripts/run_benchmark.sh"]
-
-FROM cuvs-bench AS cuvs-bench-datasets
-
-SHELL ["/bin/bash", "-euo", "pipefail", "-c"]
-
-COPY cuvs-bench/get_datasets.sh /home/rapids/cuvs-bench/get_datasets.sh
-
-COPY cuvs-bench/run_benchmarks_preloaded_datasets.sh /data/scripts/run_benchmarks_preloaded_datasets.sh
-
-RUN /home/rapids/cuvs-bench/get_datasets.sh
-
-CMD ["--dataset fashion-mnist-784-euclidean", "", "--algorithms hnswlib", ""]
-
-ENTRYPOINT ["/bin/bash", "/data/scripts/run_benchmarks_preloaded_datasets.sh"]