Merge branch 'main' into tp-cp-compose

Author: JingyaHuang

.ai/skills/diffusers-cli/run.md

@@ -108,7 +108,7 @@ Each entry calls `pipeline.load_lora_weights(<lora_id>, adapter_name=<name>)`. A
 - `--cpu-offload {model, group}` — `model` uses `enable_model_cpu_offload`, `group` uses
   `enable_group_offload(offload_type="leaf_level", use_stream=True)`. Use `group` to fit a 9B+ model on a single
   A100. Onload target device comes from `--device-map` (must be a plain device string in this case).
-- `--attention-backend {default, flash_hub, flash_varlen_hub, flash_4_hub, sage_hub}` — hub-hosted kernels,
+- `--attention-backend {default, flash_hub, flash_varlen_hub, flash_4_hub, sage_hub, sage_blackwell_hub}` — hub-hosted kernels,
   auto-downloaded on first use. Failures (kernel not available, CUDA arch mismatch, network) raise a clear
   `SystemExit` listing the alternatives instead of silently reverting to the default. Only supported on
   transformer-based pipelines; UNet pipelines get a `logger.warning` and the flag is ignored.
@@ -151,8 +151,8 @@ user also explicitly asked for a local target via `--output`.
 ## Remote execution (`--remote`)
 
 Add `--remote` to run the same call inside a [Hugging Face Sandbox](https://huggingface.co/docs/huggingface_hub/en/guides/sandbox)
-— an isolated cloud VM (built on HF Jobs) the CLI drives over HTTP: it uploads inputs, installs deps, runs
-the pipeline, downloads outputs, then terminates the sandbox.
+— an isolated cloud VM (built on HF Jobs) the CLI drives over HTTP: it uploads inputs, runs the pipeline,
+downloads outputs, then terminates the sandbox.
 
 ```bash
 diffusers-cli run \
@@ -167,12 +167,14 @@ What happens:
 
 1. Your HF token is picked up (from `--token` or your login) and forwarded into the sandbox as `HF_TOKEN`.
 2. `--pipeline-kwargs` are parsed locally so JSON errors fail fast (no wasted sandbox time).
-3. A dedicated sandbox is created on `--flavor` from a pytorch image
-   (`pytorch/pytorch:2.10.0-cuda12.8-cudnn9-runtime` by default) that already has torch + CUDA. Any local
-   file paths in `--pipeline-kwargs` are uploaded into the sandbox under `/tmp/diffusers-cli/inputs/<run_id>/`
-   via native file transfer (no bucket), and the JSON paths are rewritten to point at them.
-4. The small Python deps (`diffusers`, `accelerate`, `transformers`, `safetensors`, `sentencepiece`, `ftfy`)
-   are installed with `uv pip install --system`. Output (install + run) streams live to your terminal.
+3. A dedicated sandbox is created on `--flavor` from `diffusers/diffusers-cli-cuda:latest` by default — a
+   prebuilt image (`docker/diffusers-cli-cuda/` in this repo, rebuilt nightly) that already ships torch,
+   CUDA, `diffusers`, and the rest of the CLI's deps. Any local file paths in `--pipeline-kwargs` are
+   uploaded into the sandbox under `/tmp/diffusers-cli/inputs/<run_id>/` via native file transfer (no
+   bucket), and the JSON paths are rewritten to point at them.
+4. No dependency install runs on the default image — that step is skipped unless you passed `--dependencies`
+   (only the extras are installed then) or pointed `--image` somewhere else (the full set is installed).
+   Output streams live to your terminal.
 5. The sandbox CLI writes outputs to `/tmp/diffusers-cli/outputs/`; the CLI downloads every artifact back into
    the local target (see [`--push-to`](#-push-to) for when the download is skipped).
 6. The sandbox is terminated (unless `--keep-alive`/`--sandbox-id`), and the wallclock `run_seconds` for the
@@ -182,12 +184,13 @@ Flags:
 
 - `--flavor <name>` — sandbox hardware (e.g. `a10g-small`, `a100-large`, `4xa100-large`).
 - `--timeout <duration>` — max wallclock for the run command inside the sandbox (e.g. `30m`, `2h`). Defaults to `10m`.
-- `--dependencies <pkg>` — extra pip deps (repeatable). Appends to the defaults.
+- `--dependencies <pkg>` — extra pip deps (repeatable), installed on top of whatever the image ships.
 - `--namespace <name>` — create the sandbox under a different account.
 - `--push-to <bucket>` — see [`--push-to`](#-push-to) above. The upload runs inside the sandbox; an explicit
   value with no `--output` makes the bucket the sole destination and skips the local download.
-- `--image <ref>` — override the sandbox image. Must ship torch + CUDA; the CLI installs the small Python
-  deps on top via `uv pip install --system`. Useful for pinning a specific torch or bundling extra system libs.
+- `--image <ref>` — override the sandbox image. Must ship torch + CUDA; the CLI then installs the small
+  Python deps on top via `uv pip install --system` on every cold sandbox, which the default image avoids.
+  Useful for pinning a specific torch or bundling extra system libs.
 - `--volume <bucket-id>[:<mount-path>]` — mount an HF storage bucket into the sandbox as a read-write directory.
   Repeatable. Default mount is `/mnt/buckets/<bucket-id>`. Reference mounted files from `--pipeline-kwargs`
   like any other local path — no upload happens, the container reads straight from the FUSE mount. Applied

.github/workflows/benchmark.yml

@@ -5,8 +5,6 @@ on:
   schedule:
     - cron: "30 1 1,15 * *" # every 2 weeks on the 1st and the 15th of every month at 1:30 AM
 
-permissions:
-  contents: read
 
 env:
   DIFFUSERS_IS_CI: yes
@@ -16,8 +14,12 @@ env:
   MKL_NUM_THREADS: 8
   BASE_PATH: benchmark_outputs
 
+permissions: {}
+
 jobs:
   torch_models_cuda_benchmark_tests:
+    permissions:
+      contents: read
     env:
       SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL_BENCHMARK }}
     name: Torch Core Models CUDA Benchmarking Tests

.github/workflows/bot_pytest.yml

@@ -75,7 +75,7 @@ jobs:
         shell: bash
     steps:
       - name: Checkout PR head
-        uses: actions/checkout@v6
+        uses: actions/checkout@d23441a48e516b6c34aea4fa41551a30e30af803 # v6.1.0
         with:
           # Works for forks too — no fork credentials needed.
           ref: refs/pull/${{ github.event.issue.number }}/head
@@ -116,7 +116,7 @@ jobs:
 
       - name: Test suite reports artifacts
         if: ${{ always() }}
-        uses: actions/upload-artifact@v6
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f # v6.0.0
         with:
           name: bot_gpu_test_reports
           path: reports

.github/workflows/bot_review.yml

@@ -27,8 +27,11 @@ on:
       SERGE_INSTALLATION_ID:
         required: true
 
+permissions: {}
+
 jobs:
   relay:
+    permissions: {}
     name: Relay to Serge
     concurrency:
       group: diffusers-bot-review-${{ github.event.issue.number || github.event.pull_request.number }}

.github/workflows/build_docker_images.yml

@@ -31,7 +31,7 @@ jobs:
       pull-requests: read
     steps:
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@8d2750c68a42422c14e847fe6c8ac0403b4cbd6f  # v3
+        uses: docker/setup-buildx-action@8d2750c68a42422c14e847fe6c8ac0403b4cbd6f  # v3.12.0
 
       - name: Check out code
         uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6.0.2
@@ -54,6 +54,7 @@ jobs:
             diffusers-pytorch-xformers-cuda
             diffusers-pytorch-minimum-cuda
             diffusers-doc-builder
+            diffusers-cli-cuda
           )
 
           declare -A IMAGES_TO_BUILD=()
@@ -102,19 +103,20 @@ jobs:
           - diffusers-pytorch-xformers-cuda
           - diffusers-pytorch-minimum-cuda
           - diffusers-doc-builder
+          - diffusers-cli-cuda
 
     steps:
       - name: Checkout repository
         uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6.0.2
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@8d2750c68a42422c14e847fe6c8ac0403b4cbd6f  # v3
+        uses: docker/setup-buildx-action@8d2750c68a42422c14e847fe6c8ac0403b4cbd6f  # v3.12.0
       - name: Login to Docker Hub
-        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9  # v3
+        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9  # v3.7.0
         with:
           username: ${{ env.REGISTRY }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
       - name: Build and push
-        uses: docker/build-push-action@10e90e3645eae34f1e60eeb005ba3a3d33f178e8  # v6
+        uses: docker/build-push-action@10e90e3645eae34f1e60eeb005ba3a3d33f178e8  # v6.19.2
         with:
           no-cache: true
           context: ./docker/${{ matrix.image-name }}

.github/workflows/build_pr_documentation.yml

@@ -16,6 +16,8 @@ permissions:
 
 jobs:
   check-links:
+    permissions:
+      contents: read
     runs-on: ubuntu-latest
 
     steps:

.github/workflows/claude_review.yml

@@ -6,13 +6,15 @@ on:
   pull_request_review_comment:
     types: [created]
 
-permissions:
-  contents: write
-  pull-requests: write
-  issues: read
+
+permissions: {}
 
 jobs:
   claude-review:
+    permissions:
+      contents: write
+      issues: read
+      pull-requests: write
     if: |
       (
         github.event_name == 'issue_comment' &&

.github/workflows/issue_labeler.yml

@@ -4,12 +4,14 @@ on:
   issues:
     types: [opened]
 
-permissions:
-  contents: read
-  issues: write
+
+permissions: {}
 
 jobs:
   label:
+    permissions:
+      contents: read
+      issues: write
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd  # v6.0.2

.github/workflows/mirror_community_pipeline.yml

@@ -20,11 +20,13 @@ on:
         required: true
         default: 'main'
 
-permissions:
-  contents: read
+
+permissions: {}
 
 jobs:
   mirror_community_pipeline:
+    permissions:
+      contents: read
     env:
       SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL_COMMUNITY_MIRROR }}
     runs-on: ubuntu-22.04
@@ -69,13 +71,13 @@ jobs:
         run: |
           echo "CHECKOUT_REF: ${{ env.CHECKOUT_REF }}"
           echo "PATH_IN_REPO: ${{ env.PATH_IN_REPO }}"
-      - uses: actions/checkout@v6
+      - uses: actions/checkout@d23441a48e516b6c34aea4fa41551a30e30af803 # v6.1.0
         with:
           ref: ${{ env.CHECKOUT_REF }}
 
       # Setup + install dependencies
       - name: Set up Python
-        uses: actions/setup-python@v6
+        uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1 # v6.3.0
         with:
           python-version: "3.10"
       - name: Install dependencies

.github/workflows/nightly_tests.yml

@@ -5,8 +5,6 @@ on:
   schedule:
     - cron: "0 0 * * *" # every day at midnight
 
-permissions:
-  contents: read
 
 env:
   DIFFUSERS_IS_CI: yes
@@ -25,8 +23,12 @@ env:
   #     (torchvision::nms). The pinned set is (re)written into the override file per job below.
   UV_OVERRIDE: /tmp/uv-overrides.txt
 
+permissions: {}
+
 jobs:
   setup_torch_cuda_pipeline_matrix:
+    permissions:
+      contents: read
     name: Setup Torch Pipelines CUDA Slow Tests Matrix
     runs-on:
       group: aws-general-8-plus
@@ -58,6 +60,8 @@ jobs:
           path: reports
 
   run_nightly_tests_for_torch_pipelines:
+    permissions:
+      contents: read
     name: Nightly Torch Pipelines CUDA Tests
     needs: setup_torch_cuda_pipeline_matrix
     strategy:
@@ -110,6 +114,8 @@ jobs:
           path: reports
 
   run_nightly_tests_for_other_torch_modules:
+    permissions:
+      contents: read
     name: Nightly Torch CUDA Tests
     runs-on:
       group: aws-g4dn-2xlarge
@@ -180,6 +186,8 @@ jobs:
         path: reports
 
   run_torch_compile_tests:
+    permissions:
+      contents: read
     name: PyTorch Compile CUDA tests
 
     runs-on:
@@ -223,6 +231,8 @@ jobs:
         path: reports
 
   run_big_gpu_torch_tests:
+    permissions:
+      contents: read
     name: Torch tests on big GPU
     strategy:
       fail-fast: false
@@ -275,6 +285,8 @@ jobs:
           path: reports
 
   torch_minimum_version_cuda_tests:
+    permissions:
+      contents: read
     name: Torch Minimum Version CUDA Tests
     runs-on:
       group: aws-g4dn-2xlarge
@@ -332,6 +344,8 @@ jobs:
           path: reports
 
   run_nightly_quantization_tests:
+    permissions:
+      contents: read
     name: Torch quantization nightly tests
     strategy:
       fail-fast: false
@@ -402,6 +416,8 @@ jobs:
           path: reports
 
   run_nightly_pipeline_level_quantization_tests:
+    permissions:
+      contents: read
     name: Torch quantization nightly tests
     strategy:
       fail-fast: false
@@ -453,6 +469,8 @@ jobs:
           path: reports
 
   generate_consolidated_report:
+    permissions:
+      contents: read
     name: Generate Consolidated Test Report
     needs: [
       run_nightly_tests_for_torch_pipelines,