Add Moebius pretrained checkpoint as WebGPU-safe ONNX

UNet re-exported from hustvl/Moebius pretrained (cd01f47) with Conv3d->Conv2d; VAE graphs copied unchanged from simonw/Moebius-ONNX (5bf1ef5).

LICENSE

@@ -0,0 +1,199 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to the Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by the Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding any notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. Please also get an
+      "Alarm or alarm" page at http://www.apache.org/
+
+   Copyright 2024 Moebius Authors
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

README.md

@@ -0,0 +1,87 @@
+---
+license: apache-2.0
+library_name: onnx
+pipeline_tag: image-to-image
+base_model: hustvl/Moebius
+tags:
+  - image-inpainting
+  - inpainting
+  - object-removal
+  - diffusion
+  - onnx
+  - onnxruntime-web
+  - webgpu
+---
+
+# Moebius (pretrained) — ONNX for WebGPU
+
+ONNX exports of [Moebius](https://github.com/hustvl/Moebius) (hustvl, ECCV'26), a 0.22B-parameter
+latent diffusion model for removing things from images, for in-browser inference with
+[ONNX Runtime Web](https://onnxruntime.ai/docs/tutorials/web/) on WebGPU. Used by
+[Card Anvil](https://github.com/Card-Anvil)'s watermark remover.
+
+This repository carries the **general `pretrained` checkpoint** — Moebius before it was fine-tuned on
+any one benchmark — rather than the Places2 fine-tune that
+[simonw/Moebius-ONNX](https://huggingface.co/simonw/Moebius-ONNX) exports. The pretrained model was
+distilled from [PixelHacker](https://github.com/hustvl/PixelHacker) over its 14M image–mask
+pairs; the fine-tunes narrow it to one benchmark's photos.
+
+## Files
+
+| File               | Graph                                     | Input → output                                                                     | Size   |
+| ------------------ | ----------------------------------------- | ---------------------------------------------------------------------------------- | ------ |
+| `unet.onnx`        | Student denoiser, `pretrained` checkpoint | `latent (B,9,64,64)`, `timesteps (B,)`, `input_ids (B,10)` → `noise (B,4,64,64)`   | 907 MB |
+| `vae_encoder.onnx` | VAE encoder                               | `image (B,3,512,512)` → `moments (B,8,64,64)`                                      | 137 MB |
+| `vae_decoder.onnx` | VAE decoder                               | `latent (B,4,64,64)` → `image (B,3,512,512)`                                       | 198 MB |
+
+The UNet's inputs and outputs are identical to simonw/Moebius-ONNX's, so it drops into any pipeline
+written for that export. The two VAE graphs are copied unchanged from
+[simonw/Moebius-ONNX at `5bf1ef5`](https://huggingface.co/simonw/Moebius-ONNX/tree/5bf1ef5d2861ec01a727183a3f95dc64f352120e)
+(the VAE is [PixelHacker's](https://huggingface.co/hustvl/PixelHacker/tree/main/vae)); all four
+Moebius checkpoints share it.
+
+## How the UNet was made
+
+`export_unet.py` (here) is simonw's
+[`reexport_unet_no_conv3d.py`](https://github.com/simonw/moebius-web/blob/main/python/reexport_unet_no_conv3d.py)
+pointed at `pretrained/diffusion_pytorch_model.bin` from
+[hustvl/Moebius at `cd01f47`](https://huggingface.co/hustvl/Moebius/tree/cd01f47fb648219d3fa605806c5ce00b713faa5e).
+Each self-attention `pos_conv` Conv3d is replaced by its equivalent Conv2d so the graph compiles
+on every WebGPU backend, then the model is exported with opset 18 at a static 512 × 512
+(64 × 64 latent) — its cross-attention's position embedding is tied to that size.
+
+Run on CPU (PyTorch 2.7.1), with upstream's CUDA-only teacher import (`fla`) skipped:
+
+- Conv3d → Conv2d swap: max |Δ| 1.8e-6 against the unpatched model
+- ONNX Runtime (CPU) against PyTorch: max |Δ| 3.1e-6
+- 0 three-dimensional convolutions left in the graph
+
+## Using it
+
+The sampling loop runs outside the graphs; it must match Moebius's:
+
+- VAE scaling factor **0.13025** (not Stable Diffusion's 0.18215): encode as
+  `mean(moments[:, :4]) × 0.13025`, decode `latent / 0.13025`.
+- UNet input: `concat([noisy latent (4), mask (1), masked image's latent (4)])`.
+- `input_ids`: rows `[0…9]` conditioned, `[10…19]` unconditioned, for classifier-free guidance
+  (2 to 2.5 per the paper).
+- DDIM: `scaled_linear` betas from 0.00085 to 0.012, 1000 training steps, no clipping; 20 steps at
+  strength 0.99 run 19.
+
+## Licence
+
+Apache-2.0, as is [Moebius's code and weights](https://github.com/hustvl/Moebius/blob/main/LICENSE)
+and simonw/Moebius-ONNX. (hustvl's Hugging Face cards for Moebius and PixelHacker say MIT; their
+GitHub repositories say Apache-2.0.) The files here are converted from those works: the UNet
+re-exported as described above, the VAE graphs unchanged.
+
+## Citation
+
+```bibtex
+@inproceedings{DuanAndXu2026Moebius,
+  title     = {Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance},
+  author    = {Kangsheng Duan and Ziyang Xu and Wenyu Liu and Xiaohu Ruan and Xiaoxin Chen and Xinggang Wang},
+  year      = {2026},
+  booktitle = {European Conference on Computer Vision (ECCV)}
+}
+```

export_unet.py

@@ -0,0 +1,128 @@
+"""Export a Moebius UNet checkpoint to the WebGPU-safe ONNX graph Card Anvil loads.
+
+Adapted from simonw/moebius-web python/reexport_unet_no_conv3d.py: the
+self-attention pos_conv Conv3d is swapped for the equivalent Conv2d (so the
+graph compiles on every WebGPU backend), then exported at a fixed 512 x 512
+(64 x 64 latent) with the same input/output names as the hosted unet.onnx.
+
+usage: python export_unet.py <moebius repo> <checkpoint .bin> <out .onnx>
+"""
+import importlib.util
+import os
+import sys
+
+import numpy as np
+import torch
+from torch import nn
+
+repo, weights, out = (os.path.abspath(p) for p in sys.argv[1:4])
+sys.path.insert(0, repo)
+
+
+class PosConv2d(nn.Module):
+    """Drop-in for the Conv3d pos_conv; keeps the 5D (b, 1|dim_k, v, hh, ww) layout."""
+
+    def __init__(self, c3: nn.Conv3d):
+        super().__init__()
+        dim_k = c3.out_channels
+        r = c3.kernel_size[1]
+        self.conv = nn.Conv2d(1, dim_k, r, padding=r // 2)
+        with torch.no_grad():
+            self.conv.weight.copy_(c3.weight.squeeze(2))
+            self.conv.bias.copy_(c3.bias)
+
+    def forward(self, V):
+        b, u, v, hh, ww = V.shape
+        x = V.reshape(b * v, 1, hh, ww)
+        x = self.conv(x)
+        return x.reshape(b, v, -1, hh, ww).permute(0, 2, 1, 3, 4)
+
+
+class UNetExport(nn.Module):
+    def __init__(self, removal_model):
+        super().__init__()
+        self.m = removal_model
+
+    def forward(self, latent_in9, timesteps, input_ids):
+        return self.m(latent_in9, timesteps, input_ids).sample
+
+
+def load_removal_module():
+    # Straight from the file: removal/v1_2/__init__.py drags in the training
+    # pipeline and dataset (cv2, transformers, ...), none of which export needs.
+    path = os.path.join(repo, "removal", "v1_2", "removal_model.py")
+    spec = importlib.util.spec_from_file_location("removal_model", path)
+    module = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(module)
+    return module
+
+
+def main():
+    rm = load_removal_module()
+    cfg = rm.load_cfg(os.path.join(repo, "config", "model_cfg", "moebius.yaml"))
+    model = rm.build_removal_model(cfg, 20).eval()
+    state = torch.load(weights, map_location="cpu", weights_only=True)
+    print("[load]", model.load_state_dict(state, strict=True))
+    print(f"[load] params = {sum(p.numel() for p in model.parameters()) / 1e6:.2f}M")
+
+    torch.manual_seed(0)
+    latent9 = torch.randn(2, 9, 64, 64)
+    ts = torch.tensor([811, 811], dtype=torch.int64)
+    ids = torch.tensor([list(range(10, 20)), list(range(10))], dtype=torch.int64)
+
+    wrap = UNetExport(model)
+    with torch.no_grad():
+        before = wrap(latent9, ts, ids).numpy()
+    n = 0
+    for mod in model.modules():
+        pc = getattr(mod, "pos_conv", None)
+        if isinstance(pc, nn.Conv3d):
+            mod.pos_conv = PosConv2d(pc)
+            n += 1
+    with torch.no_grad():
+        after = wrap(latent9, ts, ids).numpy()
+    d = np.abs(before - after).max()
+    print(f"[fix] replaced {n} Conv3d -> Conv2d; torch max|diff| = {d:.3e}")
+    assert n > 0 and d < 1e-4, "Conv2d replacement changed the math"
+
+    os.makedirs(os.path.dirname(out), exist_ok=True)
+    torch.onnx.export(
+        wrap,
+        (latent9, ts, ids),
+        out,
+        opset_version=18,
+        dynamo=False,
+        input_names=["latent", "timesteps", "input_ids"],
+        output_names=["noise"],
+        dynamic_axes={
+            "latent": {0: "B"},
+            "timesteps": {0: "B"},
+            "input_ids": {0: "B"},
+            "noise": {0: "B"},
+        },
+    )
+
+    import onnx
+    import onnxruntime as ort
+
+    sess = ort.InferenceSession(out, providers=["CPUExecutionProvider"])
+    got = sess.run(
+        None,
+        {"latent": latent9.numpy(), "timesteps": ts.numpy(), "input_ids": ids.numpy()},
+    )[0]
+    print(f"[check] ONNX vs torch max|diff| = {np.abs(after - got).max():.3e}")
+    for i in sess.get_inputs() + sess.get_outputs():
+        print(f"[check] {i.name}: {i.type} {i.shape}")
+    graph = onnx.load(out, load_external_data=False)
+    conv3d = sum(
+        1
+        for node in graph.graph.node
+        if node.op_type == "Conv"
+        and any(a.name == "kernel_shape" and len(a.ints) == 3 for a in node.attribute)
+    )
+    print(f"[check] 3D Conv nodes left: {conv3d}")
+    print(f"[done] {out} ({os.path.getsize(out):,} bytes)")
+
+
+if __name__ == "__main__":
+    main()

unet.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2ea6bed5156872091294ae238d127a4e2987dfea288321341cca59061cb1549
+size 906698977

vae_decoder.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d90ef0b7f6c8c8b7234459c8b449d70be0033bf1576c842e8b9991baf3934280
+size 198078671

vae_encoder.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8b81d41e757222a0707665ba9d826703987855e5bed056036b90b988968042f
+size 136757093