Merge 0df0b0d613 into a3020f107e

* fix(Video): don't crash on videos with undecodable audio streams

Signed-off-by: bigcat88 <bigcat88@icloud.com>

* Update comfy_api_nodes/util/upload_helpers.py

---------

Signed-off-by: bigcat88 <bigcat88@icloud.com>
Co-authored-by: Alexis Rolland <alexisrolland@hotmail.com>

.github/workflows/cla.yml

@@ -0,0 +1,91 @@
+name: CLA Assistant
+
+on:
+  issue_comment:
+    types: [created]
+  pull_request_target:
+    types: [opened, synchronize, closed]
+
+permissions:
+  actions: write
+  contents: read # 'read' is enough because signatures live in a REMOTE repo
+  pull-requests: write
+  statuses: write
+
+jobs:
+  cla-assistant:
+    runs-on: ubuntu-latest
+    steps:
+      # The CLA action normally requires every commit author in a PR to sign.
+      # We only want the PR author to sign, so we allowlist all other committers
+      # by computing them from the PR's commits and excluding the PR author.
+      - name: Build author-only allowlist
+        id: allowlist
+        if: >
+          github.event_name == 'pull_request_target' ||
+          (github.event_name == 'issue_comment' && github.event.issue.pull_request && (
+            github.event.comment.body == 'recheck' ||
+            github.event.comment.body == 'I have read and agree to the Contributor License Agreement'
+          ))
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          PR_NUMBER: ${{ github.event.pull_request.number || github.event.issue.number }}
+          PR_AUTHOR: ${{ github.event.pull_request.user.login || github.event.issue.user.login }}
+          BASE_ALLOWLIST: action@github.com,actions-user,ampagent,claude,comfy-pr-bot,GitHub Action,github-actions,github-actions[bot],Glary Bot,Glary-Bot,*[bot]
+        run: |
+          others=$(gh api "repos/${{ github.repository }}/pulls/${PR_NUMBER}/commits" --paginate \
+            --jq '.[] | (.author.login // empty), (.committer.login // empty)' \
+            | sort -u | grep -vix "${PR_AUTHOR}" | paste -sd, -)
+          if [ -n "$others" ]; then
+            echo "allowlist=${BASE_ALLOWLIST},${others}" >> "$GITHUB_OUTPUT"
+          else
+            echo "allowlist=${BASE_ALLOWLIST}" >> "$GITHUB_OUTPUT"
+          fi
+
+      - name: CLA Assistant
+        # Run on PR events, on "recheck" comment, or when someone posts the exact signing phrase.
+        # IMPORTANT: this phrase must match `custom-pr-sign-comment` below.
+        if: >
+          github.event_name == 'pull_request_target' ||
+          (github.event_name == 'issue_comment' && github.event.issue.pull_request && (
+            github.event.comment.body == 'recheck' ||
+            github.event.comment.body == 'I have read and agree to the Contributor License Agreement'
+          ))
+        uses: contributor-assistant/github-action@ca4a40a7d1004f18d9960b404b97e5f30a505a08 # v2.6.1
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          # PAT required to write to the centralized signatures repo.
+          PERSONAL_ACCESS_TOKEN: ${{ secrets.PERSONAL_ACCESS_TOKEN }}
+        with:
+          # Where the CLA document lives (shown to contributors)
+          path-to-document: https://github.com/Comfy-Org/comfy-cla/blob/main/comfyui_icla.md
+
+          # Centralized signature storage
+          remote-organization-name: comfy-org
+          remote-repository-name: comfy-cla
+          path-to-signatures: signatures/cla.json
+          branch: main
+
+          # Only the PR author must sign: bots plus every non-author committer
+          # are allowlisted via the "Build author-only allowlist" step above.
+          # *[bot] is a catch-all for any GitHub App bot account.
+          allowlist: ${{ steps.allowlist.outputs.allowlist }}
+
+          # Custom PR comment messages
+          custom-notsigned-prcomment: |
+            🎉 Thank you for your contribution, we really appreciate it! 🎉
+
+            Like many open source projects, we require contributors to sign our [Contributor License Agreement (CLA)](https://github.com/Comfy-Org/comfy-cla/blob/main/comfyui_icla.md). A CLA makes the ownership of contributions explicit, so contributors and the project share a clear understanding of how the code can be used. By signing, you:
+
+            - Confirm that you own your contribution.
+            - Keep the right to reuse your own code.
+            - Grant us a copyright license to include and share it within our projects.
+
+            CLAs are standard practice across major open source projects including those under the Apache Software Foundation and the Linux Foundation. Ours is based on the Apache Software Foundation's CLA. Most importantly, it would enable us to relicense the project under a more permissive license in the future, giving the project and its community greater flexibility.
+
+            ✍ **To sign, please post a new comment on this PR with exactly the following text:** ✍
+
+          custom-pr-sign-comment: I have read and agree to the Contributor License Agreement
+
+          custom-allsigned-prcomment: |
+            ✅ All contributors have signed the CLA. Thank you! This PR is ready to be merged.

CLAUDE.md

@@ -1 +0,0 @@
-AGENTS.md

baselines/environment_rocm_working.txt

@@ -0,0 +1,83 @@
+aiohappyeyeballs==2.6.2
+aiohttp==3.14.1
+aiosignal==1.4.0
+alembic==1.18.4
+annotated-doc==0.0.4
+annotated-types==0.7.0
+anyio==4.13.0
+attrs==26.1.0
+av==17.1.0
+blake3==1.0.8
+certifi==2026.5.20
+charset-normalizer==3.4.7
+click==8.4.1
+colorama==0.4.6
+comfy-aimdo==0.4.9
+comfy-kitchen==0.2.10
+comfyui-embedded-docs==0.5.3
+comfyui-workflow-templates-core==0.3.252
+comfyui-workflow-templates-media-api==0.3.80
+comfyui-workflow-templates-media-image==0.3.150
+comfyui-workflow-templates-media-other==0.3.217
+comfyui-workflow-templates-media-video==0.3.91
+comfyui_frontend_package==1.45.15
+comfyui_workflow_templates==0.9.98
+einops==0.8.2
+filelock==3.29.4
+frozenlist==1.8.0
+fsspec==2026.4.0
+glfw==2.10.0
+greenlet==3.5.1
+h11==0.16.0
+hf-xet==1.5.1
+httpcore==1.0.9
+httpx==0.28.1
+huggingface_hub==1.19.0
+idna==3.18
+Jinja2==3.1.6
+kornia==0.8.3
+kornia_rs==0.1.14
+Mako==1.3.12
+markdown-it-py==4.2.0
+MarkupSafe==3.0.3
+mdurl==0.1.2
+mpmath==1.3.0
+multidict==6.7.1
+networkx==3.6.1
+numpy==1.26.4
+packaging==26.2
+pillow==12.2.0
+propcache==0.5.2
+psutil==7.2.2
+pydantic==2.13.4
+pydantic-settings==2.14.1
+pydantic_core==2.46.4
+Pygments==2.20.0
+PyOpenGL==3.1.10
+python-dotenv==1.2.2
+PyYAML==6.0.3
+regex==2026.5.9
+requests==2.34.2
+rich==15.0.0
+safetensors==0.8.0
+scipy==1.17.1
+sentencepiece==0.2.1
+setuptools==82.0.1
+shellingham==1.5.4
+simpleeval==1.0.7
+spandrel==0.4.2
+SQLAlchemy==2.0.50
+sympy==1.14.0
+tokenizers==0.22.2
+torch @ https://github.com/scottt/rocm-TheRock/releases/download/v6.5.0rc-pytorch/torch-2.7.0a0+git3f903c3-cp312-cp312-win_amd64.whl#sha256=ab308d20b8568354781ceaad1c9a1637b6dff16ab42e589fa87b19fa87f3c839
+torchaudio @ https://github.com/scottt/rocm-TheRock/releases/download/v6.5.0rc-pytorch/torchaudio-2.6.0a0+1a8f621-cp312-cp312-win_amd64.whl#sha256=caa1291b5040325d67ac2d6bddb9c3ec9478337dfc70a4d08bda8a557c834698
+torchsde==0.2.6
+torchvision @ https://github.com/scottt/rocm-TheRock/releases/download/v6.5.0rc-pytorch/torchvision-0.22.0+9eb57cd-cp312-cp312-win_amd64.whl#sha256=47fbcdc9b5e80ee7ab40c27bbf5cd36f7a7091eae3d43a09eebd833a391de1ec
+tqdm==4.68.2
+trampoline==0.1.2
+transformers==5.12.0
+typer==0.25.1
+typing-inspection==0.4.2
+typing_extensions==4.15.0
+urllib3==2.7.0
+yarl==1.24.2

baselines/system_info.txt

@@ -0,0 +1,36 @@
+=== ComfyUI ROCm Stable Baseline ===
+Date: 2026-06-20
+
+--- Python ---
+Python: 3.12.10
+
+--- PyTorch / ROCm ---
+torch:      2.7.0a0+git3f903c3
+CUDA avail: True
+Device:     AMD Radeon(TM) 8050S Graphics
+VRAM (GB):  14.37
+ROCm/HIP:   6.5.25205-c1c2abe52
+
+--- torch packages ---
+torch: 2.7.0a0+git3f903c3
+torchvision: 0.22.0+9eb57cd
+torchaudio: 2.6.0a0+1a8f621
+
+--- ComfyUI ---
+Version: 0.24.0
+Backend: ROCm 6.5 (scottt/rocm-TheRock gfx1151 wheel)
+Tested:  SD1.5 512x512 20steps ~5 it/s, stable
+
+--- Wheel sources (gfx1151 / Strix Halo) ---
+torch:       scottt/rocm-TheRock v6.5.0rc-pytorch/torch-2.7.0a0+git3f903c3-cp312-cp312-win_amd64.whl
+torchvision: scottt/rocm-TheRock v6.5.0rc-pytorch/torchvision-0.22.0+9eb57cd-cp312-cp312-win_amd64.whl
+torchaudio:  scottt/rocm-TheRock v6.5.0rc-pytorch/torchaudio-2.6.0a0+1a8f621-cp312-cp312-win_amd64.whl
+numpy:       pinned to <2 (1.26.4) for wheel compatibility
+
+--- Startup ---
+cd %USERPROFILE%\ComfyUI
+.\\venv\\Scripts\\activate
+python main.py
+
+--- Saved workflow ---
+baselines/workflows/sd15_test_rocm_workflow.json

baselines/system_info_realvisxl.txt

@@ -0,0 +1,55 @@
+=== ComfyUI RealVisXL no-mmap Stable Baseline ===
+Date: 2026-06-20
+
+--- Previous baseline ---
+Tag: rocm-sd15-working-baseline (preserved, not modified)
+
+--- This baseline adds ---
+Fix:   comfy/utils.py: _load_safetensors_no_mmap() for files >4 GB
+Model: RealVisXL_V4.0.safetensors (6.46 GB) - path only, not in git
+Test:  768x1024, 25 steps, cfg=6, dpmpp_2m, karras -> OK
+
+--- Root cause of crash (diagnosed & fixed) ---
+Strix Halo UMA: ROCm init reserves ~14 GB GPU virtual address space.
+safetensors mmap of files >~4 GB then fails (Windows VA space exhausted).
+SD1.5 (3.97 GB) < threshold -> mmap OK.
+SDXL fp16 (~6.5 GB) > threshold -> access violation in safe_open().
+Fix: sequential file-read (open+seek+read) bypasses mmap entirely.
+
+--- Patch location ---
+File:      comfy/utils.py
+Functions: _load_safetensors_no_mmap(), _LARGE_FILE_MMAP_THRESHOLD = 4_000_000_000
+Branch:    load_torch_file() elif os.path.getsize > threshold and cuda available
+
+--- Startup command ---
+cd %USERPROFILE%\ComfyUI
+.\venv\Scripts\activate
+python main.py --disable-dynamic-vram --disable-mmap
+
+--- GPU / ROCm ---
+torch:    2.7.0a0+git3f903c3
+Device:   AMD Radeon(TM) 8050S Graphics
+VRAM GB:  14.37
+ROCm:     6.5 / gfx1151 (Strix Halo)
+
+--- Models in checkpoints (not in git) ---
+v1-5-pruned-emaonly.safetensors   3.97 GB  SD1.5 baseline
+RealVisXL_V4.0.safetensors        6.46 GB  SDXL realistic portrait
+
+--- Working parameters (RealVisXL) ---
+Resolution:   768x1024
+Steps:        25
+CFG:          6
+Sampler:      dpmpp_2m
+Scheduler:    karras
+Batch size:   1
+
+--- Recovery commands ---
+# 1. Return to this code state:
+git checkout rocm-realvisxl-nommap-working
+
+# 2. Re-download RealVisXL if needed (not in git):
+# python -c "from huggingface_hub import hf_hub_download; hf_hub_download(repo_id='SG161222/RealVisXL_V4.0', filename='RealVisXL_V4.0.safetensors', local_dir='models/checkpoints')"
+
+# 3. Start ComfyUI:
+# python main.py --disable-dynamic-vram --disable-mmap

baselines/workflows/sd15_test_rocm_workflow.json

@@ -0,0 +1,107 @@
+{
+  "2": {
+    "inputs": {
+      "ckpt_name": "v1-5-pruned-emaonly.safetensors"
+    },
+    "class_type": "CheckpointLoaderSimple",
+    "_meta": {
+      "title": "Checkpoint加载器（简易）"
+    }
+  },
+  "3": {
+    "inputs": {
+      "text": "aa cute fluffy kitten, big round eyes, detailed fur, soft natural window light, cozy indoor background, shallow depth of field, photorealistic, high quality, 50mm lens",
+      "clip": [
+        "2",
+        1
+      ]
+    },
+    "class_type": "CLIPTextEncode",
+    "_meta": {
+      "title": "CLIP文本编码"
+    }
+  },
+  "4": {
+    "inputs": {
+      "text": "low quality, blurry, deformed, ugly, bad anatomy, distorted face, extra limbs, bad eyes, oversaturated",
+      "clip": [
+        "2",
+        1
+      ]
+    },
+    "class_type": "CLIPTextEncode",
+    "_meta": {
+      "title": "CLIP文本编码"
+    }
+  },
+  "5": {
+    "inputs": {
+      "width": 512,
+      "height": 512,
+      "batch_size": 1
+    },
+    "class_type": "EmptyLatentImage",
+    "_meta": {
+      "title": "空Latent图像"
+    }
+  },
+  "6": {
+    "inputs": {
+      "seed": 826325619577598,
+      "steps": 30,
+      "cfg": 7,
+      "sampler_name": "dpmpp_2m",
+      "scheduler": "normal",
+      "denoise": 1,
+      "model": [
+        "2",
+        0
+      ],
+      "positive": [
+        "3",
+        0
+      ],
+      "negative": [
+        "4",
+        0
+      ],
+      "latent_image": [
+        "5",
+        0
+      ]
+    },
+    "class_type": "KSampler",
+    "_meta": {
+      "title": "K采样器"
+    }
+  },
+  "7": {
+    "inputs": {
+      "samples": [
+        "6",
+        0
+      ],
+      "vae": [
+        "2",
+        2
+      ]
+    },
+    "class_type": "VAEDecode",
+    "_meta": {
+      "title": "VAE解码"
+    }
+  },
+  "8": {
+    "inputs": {
+      "filename_prefix": "ComfyUI",
+      "images": [
+        "7",
+        0
+      ]
+    },
+    "class_type": "SaveImage",
+    "_meta": {
+      "title": "保存图像"
+    }
+  }
+}

comfy/utils.py

@@ -22,6 +22,7 @@ import math
 import struct
 import ctypes
 import os
+import sys
 import comfy.memory_management
 import safetensors.torch
 import numpy as np
@@ -119,6 +120,40 @@ def load_safetensors(ckpt):
     return sd, header.get("__metadata__", {}),
 
 
+_LARGE_FILE_MMAP_THRESHOLD = 4_000_000_000  # 4 GB
+
+
+def _load_safetensors_no_mmap(ckpt, device=None):
+    # Windows + ROCm/CUDA UMA: large mmaps fail after GPU virtual address space is reserved.
+    # Read tensors sequentially from file instead.
+    if device is None:
+        device = torch.device("cpu")
+    sd = {}
+    with open(ckpt, "rb") as fh:
+        header_len = struct.unpack("<Q", fh.read(8))[0]
+        header = json.loads(fh.read(header_len).decode("utf-8"))
+        data_start = 8 + header_len
+        for name, info in header.items():
+            if name == "__metadata__":
+                continue
+            start, end = info["data_offsets"]
+            dtype = _TYPES[info["dtype"]]
+            shape = info["shape"]
+            expected = end - start
+            if expected == 0:
+                sd[name] = torch.empty(shape, dtype=dtype, device=device)
+                continue
+            fh.seek(data_start + start)
+            raw = fh.read(expected)
+            if len(raw) != expected:
+                raise RuntimeError(
+                    f"Safetensors read error: tensor '{name}' expected {expected} bytes, got {len(raw)}. "
+                    f"File may be corrupt or truncated."
+                )
+            sd[name] = torch.frombuffer(bytearray(raw), dtype=dtype).reshape(shape).clone().to(device=device)
+    return sd, header.get("__metadata__", {})
+
+
 def load_torch_file(ckpt, safe_load=False, device=None, return_metadata=False):
     if device is None:
         device = torch.device("cpu")
@@ -129,6 +164,15 @@ def load_torch_file(ckpt, safe_load=False, device=None, return_metadata=False):
                 sd, metadata = load_safetensors(ckpt)
                 if not return_metadata:
                     metadata = None
+            elif (os.path.getsize(ckpt) > _LARGE_FILE_MMAP_THRESHOLD
+                  and sys.platform == "win32"
+                  and torch.cuda.is_available()):
+                # Windows ROCm/UMA: GPU init reserves ~14 GB of virtual address space,
+                # preventing mmap of files >4 GB. Use sequential file-read instead.
+                # Scoped to Windows only to avoid overhead on Linux/Mac CUDA systems.
+                sd, metadata = _load_safetensors_no_mmap(ckpt, device=device)
+                if not return_metadata:
+                    metadata = None
             else:
                 with safetensors.safe_open(ckpt, framework="pt", device=device.type) as f:
                     sd = {}

comfy_api/latest/_input_impl/video_types.py

@@ -281,11 +281,18 @@ class VideoFromFile(VideoInput):
         video_done = False
         audio_done = True
 
-        if len(container.streams.audio):
-            audio_stream = container.streams.audio[-1]
+        # Use the last decodable audio stream. Streams FFmpeg has no decoder for have no codec context,
+        # and decoding their packets crashes the process. (e.g. APAC spatial-audio track in iPhone)
+        audio_stream = next(
+            (s for s in reversed(container.streams.audio) if s.codec_context is not None),
+            None,
+        )
+        if audio_stream is not None:
             streams += [audio_stream]
             resampler = av.audio.resampler.AudioResampler(format='fltp')
             audio_done = False
+        elif len(container.streams.audio):
+            logging.warning("No decodable audio stream found in video; ignoring audio.")
 
         for packet in container.demux(*streams):
             if video_done and audio_done:
@@ -457,10 +464,13 @@ class VideoFromFile(VideoInput):
                         else:
                             output_container.metadata[key] = json.dumps(value)
 
-                # Add streams to the new container
+                # Add streams to the new container. Streams with no codec context cannot be used as an output template.
                 stream_map = {}
                 for stream in streams:
                     if isinstance(stream, (av.VideoStream, av.AudioStream, SubtitleStream)):
+                        if stream.codec_context is None:
+                            logging.warning("Skipping %s stream %d with unsupported codec", stream.type, stream.index)
+                            continue
                         out_stream = output_container.add_stream_from_template(template=stream, opaque=True)
                         stream_map[stream] = out_stream
 

comfy_api_nodes/util/upload_helpers.py

@@ -158,7 +158,14 @@ async def upload_video_to_comfyapi(
 
     # Convert VideoInput to BytesIO using specified container/codec
     video_bytes_io = BytesIO()
-    video.save_to(video_bytes_io, format=container, codec=codec)
+    try:
+        video.save_to(video_bytes_io, format=container, codec=codec)
+    except Exception as e:
+        raise ValueError(
+            f"Could not convert the input video to {container.value.upper()} for upload; "
+            f"the file may be corrupted or use an unsupported codec. "
+            f"Try re-exporting it as MP4 (H.264). Original error: {e}"
+        ) from e
     video_bytes_io.seek(0)
 
     return await upload_file_to_comfyapi(cls, video_bytes_io, filename, upload_mime_type, wait_label)