Merge 0df0b0d613 into 6880614319

- utils.py: add device param to _load_safetensors_no_mmap, move tensors
  to target device instead of always returning CPU tensors
- utils.py: validate read length == expected bytes; raise RuntimeError
  on partial/corrupt reads instead of silently creating empty tensors
- utils.py: scope no-mmap fallback to sys.platform == win32 to avoid
  unnecessary overhead on Linux/Mac CUDA systems; add sys import
- baselines: replace hardcoded LvHHu username with %USERPROFILE% in
  startup commands for portability

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

AGENTS.md

@@ -127,6 +127,8 @@
 - Do not add unnecessary `try`/`except` blocks. Use them for optional dependency,
   platform, or backend capability detection only when the program has a useful
   fallback. Prefer specific exception types when changing new code.
+- If a library version is pinned in `requirements.txt`, do not add code to
+  ComfyUI to handle older versions of that library.
 - Remove any workarounds for PyTorch versions that ComfyUI no longer officially
   supports. Deprecated workarounds include catching an exception and rerunning
   the same op with the input cast to float. If a workaround does not have a

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_extras/nodes_text_overlay.py

@@ -0,0 +1,150 @@
+import numpy as np
+import torch
+from PIL import Image as PILImage, ImageColor, ImageDraw, ImageFont
+from typing_extensions import override
+
+from comfy_api.latest import ComfyExtension, IO
+
+
+class TextOverlay(IO.ComfyNode):
+    @classmethod
+    def define_schema(cls):
+        return IO.Schema(
+            node_id="TextOverlay",
+            display_name="Draw Text Overlay",
+            category="text",
+            description="Draw text overlay on an image or batch of images.",
+            search_aliases=["text", "label", "caption", "subtitle", "watermark", "title", "addlabel", "overlay"],
+            inputs=[
+                IO.Image.Input("images"),
+                IO.String.Input("text", multiline=True, default=""),
+                IO.Float.Input("font_size", default=5.0, min=0.5, max=50.0, step=0.5, tooltip="Font size as a percentage of the image height."),
+                IO.Color.Input("color", default="#ffffff", tooltip="Color of the text."),
+                IO.Combo.Input("position", options=["top", "bottom"], default="top"),
+                IO.Combo.Input("align", options=["left", "center", "right"], default="left"),
+                IO.Boolean.Input("outline", default=True, tooltip="Draw a black outline around the text."),
+            ],
+            outputs=[IO.Image.Output(display_name="images")],
+        )
+
+    @classmethod
+    def execute(cls, images, text, font_size, color, position, align, outline) -> IO.NodeOutput:
+        if text.strip() == "":
+            return IO.NodeOutput(images)
+
+        text = text.replace("\\n", "\n").replace("\\t", "\t")
+
+        text_rgba = cls.parse_color_to_rgba(color)
+        outline_rgba = (0, 0, 0, 255) if outline else (0, 0, 0, 0)
+
+        # Render the overlay once and composite it across all frames in the batch
+        height = images.shape[1]
+        width = images.shape[2]
+        overlay_rgb, overlay_alpha = cls.render_overlay_text(width, height, text, position, align, font_size, text_rgba, outline_rgba)
+        overlay_rgb = overlay_rgb.to(device=images.device, dtype=images.dtype)
+        overlay_alpha = overlay_alpha.to(device=images.device, dtype=images.dtype)
+
+        result = images * (1.0 - overlay_alpha) + overlay_rgb * overlay_alpha
+        return IO.NodeOutput(result)
+
+    @staticmethod
+    def parse_color_to_rgba(color_string):
+        parsed = ImageColor.getrgb(color_string)
+
+        if len(parsed) == 3:
+            return (*parsed, 255)
+
+        return parsed
+
+    @classmethod
+    def render_overlay_text(cls, width, height, text, position, align, font_size, text_rgba, outline_rgba):
+        line_spacing = 1.2
+        margin_percent = 1.0
+        min_font_percent = 2.0
+        min_font_pixels = 10
+        outline_thickness_factor = 0.04
+
+        # Draw onto a transparent layer so the result can be alpha-composited over any frame.
+        layer = PILImage.new("RGBA", (width, height), (0, 0, 0, 0))
+        draw = ImageDraw.Draw(layer)
+
+        margin = int(round(margin_percent / 100.0 * min(width, height)))
+        max_width = max(1, width - 2 * margin)
+        max_height = max(1, height - 2 * margin)
+
+        # Font scales with resolution, then shrinks to fit the height.
+        size = max(1, int(round(font_size / 100.0 * height)))
+        floor = min(size, max(min_font_pixels, int(round(min_font_percent / 100.0 * height))))
+
+        while True:
+            font = ImageFont.load_default(size=size)
+            stroke = max(1, int(round(size * outline_thickness_factor))) if outline_rgba[3] > 0 else 0
+            block = "\n".join(cls.wrap_text(text, font, max_width))
+            # convert line spacing to pixel spacing
+            single = draw.textbbox((0, 0), "Ay", font=font, stroke_width=stroke)
+            double = draw.multiline_textbbox((0, 0), "Ay\nAy", font=font, spacing=0, stroke_width=stroke)
+            natural_advance = (double[3] - double[1]) - (single[3] - single[1])
+            pixel_spacing = int(round(size * line_spacing - natural_advance))
+            box = draw.multiline_textbbox((0, 0), block, font=font, spacing=pixel_spacing, stroke_width=stroke)
+            block_height = box[3] - box[1]
+
+            if block_height <= max_height or size <= floor:
+                break
+
+            size = max(floor, int(size * 0.9))
+
+        anchor_h, x = {"left": ("l", margin), "center": ("m", width / 2), "right": ("r", width - margin)}[align]
+
+        # Offset y so the rendered text sits flush against the margin
+        if position == "bottom":
+            y = height - margin - box[3]
+        else:
+            y = margin - box[1]
+
+        draw.multiline_text((x, y), block, font=font, fill=text_rgba, anchor=anchor_h + "a",
+                            align=align, spacing=pixel_spacing, stroke_width=stroke, stroke_fill=outline_rgba)
+
+        overlay = np.array(layer).astype(np.float32) / 255.0
+        overlay_rgb = torch.from_numpy(overlay[:, :, :3])
+        overlay_alpha = torch.from_numpy(overlay[:, :, 3:4])
+        return overlay_rgb, overlay_alpha
+
+    @staticmethod
+    def wrap_text(text, font, max_width):
+        lines = []
+        for raw_line in text.split("\n"):
+            words = raw_line.split()
+            if not words:
+                lines.append("")
+                continue
+            current = ""
+            # Break the line into words and split words that are too long
+            for word in words:
+                while font.getlength(word) > max_width and len(word) > 1:
+                    cut = 1
+                    while cut < len(word) and font.getlength(word[:cut + 1]) <= max_width:
+                        cut += 1
+                    if current:
+                        lines.append(current)
+                        current = ""
+                    lines.append(word[:cut])
+                    word = word[cut:]
+                candidate = word if not current else current + " " + word
+                if not current or font.getlength(candidate) <= max_width:
+                    current = candidate
+                else:
+                    lines.append(current)
+                    current = word
+            if current:
+                lines.append(current)
+        return lines
+
+
+class TextOverlayExtension(ComfyExtension):
+    @override
+    async def get_node_list(self) -> list[type[IO.ComfyNode]]:
+        return [TextOverlay]
+
+
+async def comfy_entrypoint() -> TextOverlayExtension:
+    return TextOverlayExtension()

nodes.py

@@ -2478,6 +2478,7 @@ async def init_builtin_extra_nodes():
         "nodes_glsl.py",
         "nodes_lora_debug.py",
         "nodes_textgen.py",
+        "nodes_text_overlay.py",
         "nodes_color.py",
         "nodes_toolkit.py",
         "nodes_replacements.py",