SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2108.01073 · 2021

SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations

Jiajun Wu, Yang Song, Chenlin Meng, Stefano Ermon, Jiaming Song, Jun-Yan Zhu, Yutong He

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 2 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2108.01073.json · badge: /api/ran/2108.01073.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
ermongroup/SDEdit marked official in Papers with Code
Papers with Code archive
2 ran · 0 no recorded run
FunctionStatusWhere it lives
extract Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ermongroup/SDEdit/runners/image_editing.py
code served (permissive licence) · get_code("6b36037c0def2c14")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 3, dtype family float, shape (2, 1, 1) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function expects:
# - a: array-like that can be converted to tensor and indexed with gather
# - t: tensor of shape (bs,) containing indices
# - x_shape: tuple where x_shape[0] == bs and len(x_shape) >= 1

bs = 2  # batch size
num_coeffs = 5  # number of coefficients in array a

# Create a small coefficient array
a = np.array([0.1, 0.2, 0.3, 0.4, 0.5])

# Create time indices tensor of shape (bs,)
t = torch.tensor([0, 2], dtype=torch.long)

# Create x_shape: first dimension must match bs, other dimensions arbitrary
x_shape = (2, 4, 3)  # batch_size=2, height=4, width=3

ARGS = {
    "a": a,
    "t": t,
    "x_shape": x_shape
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 1, 1) — shape (2, 1, 1) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
image_editing_denoising_step_flexible_mask Ran
ran; our contract draft was wrong, not the code
ermongroup/SDEdit/runners/image_editing.py
code served (permissive licence) · get_code("b1a01236b34cd196")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (2, 3, 4, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Define a simple model that takes x and t and returns output
class SimpleModel(nn.Module):
    def __init__(self, input_dim, output_dim):
        super().__init__()
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, x, t):
        # x shape: (bs, channels, height, width) or similar
        # t shape: (bs,)
        # Output should match x shape
        bs = x.shape[0]
        # Flatten x, process, reshape back
        x_flat = x.reshape(bs, -1)
        out = self.linear(x_flat)
        return out.reshape(x.shape)

# Set up parameters
batch_size = 2
channels = 3
spatial_dim = 4

# Create input tensor x: (batch_size, channels, spatial_dim, spatial_dim)
x = torch.randn(batch_size, channels, spatial_dim, spatial_dim)

# Create timestep tensor t: (batch_size,)
t = torch.tensor([5, 3], dtype=torch.long)

# Create model
model = SimpleModel(input_dim=channels * spatial_dim * spatial_dim, 
                    output_dim=channels * spatial_dim * spatial_dim)

# Create logvar: should be indexable by time steps, typically shape (num_timesteps,)
num_timesteps = 10
logvar = torch.randn(num_timesteps)

# Create betas: should be shape (num_timesteps,)
betas = torch.linspace(0.0001, 0.02, num_timesteps)

ARGS = {
    'x': x,
    't': t,
    'model': model,
    'logvar': logvar,
    'betas': betas,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 3, 4, 4) — shape (2, 3, 4, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 2 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Guided image synthesis enables everyday users to create and edit photo-realistic images with minimum effort. The key challenge is balancing faithfulness to the user inputs (e.g., hand-drawn colored strokes) and realism of the synthesized images. Existing GAN-based methods attempt to achieve such balance using either conditional GANs or GAN inversions, which are challenging and often require additional training data or loss functions for individual applications. To address these issues, we introduce a new image synthesis and editing method, Stochastic Differential Editing (SDEdit), based on a diffusion model generative prior, which synthesizes realistic images by iteratively denoising through a stochastic differential equation (SDE). Given an input image with user guide in a form of manipulating RGB pixels, SDEdit first adds noise to the input, then subsequently denoises the resulting image through the SDE prior to increase its realism. SDEdit does not require task-specific training or inversions and can naturally achieve the balance between realism and faithfulness. SDEdit outperforms state-of-the-art GAN-based methods by up to 98.09% on realism and 91.72% on overall satisfaction scores, according to a human perception study, on multiple tasks, including stroke-based image synthesis and editing as well as image compositing.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2108.01073")
get_code_for_paper("2108.01073")
have("2108.01073")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2108.01073.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2108.01073.svg)](https://syntology.ai/paper/2108.01073)

Connect an agent — have() is free.