SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2212.09748 · 2022

Scalable Diffusion Models with Transformers

Saining Xie, William Peebles

arXiv · PDF · Open in the Atlas

Code that ran

15 of 20 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; 2 honoured the contract we drafted, 0 violated it and 10 have no contract verdict; 3 failures of our instrument; 5 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/2212.09748.json · badge: /api/ran/2212.09748.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
milmor/diffusion-transformer-keras community (classifier label)
classifier basis not recorded on the link
6 ran · 0 no recorded run
milmor/diffusion-transformer community (classifier label)
classifier basis: plan_row
5 ran · 2 no recorded run
copy not recorded community
not in the paper's declared repositories
2 ran · 2 no recorded run
VachanVY/diffusion-transformer reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 1 no recorded run
hustvl/dig extension (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_activation_statistics Ran
ran on a synthesized input
milmor/diffusion-transformer-keras/fid.py
code served (permissive licence) · get_code("55d54369eada8946")
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 tuple. The image it ran in was not recorded. If your copy returns the same type 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 numpy as np
import tensorflow as tf

# Create a mock Inception model that outputs 2048-dimensional activations
class MockInception(tf.Module):
    def __call__(self, batch):
        # batch shape: (batch_size, height, width, channels)
        # Return shape: (batch_size, 2048)
        batch_size = tf.shape(batch)[0]
        return tf.random.normal((batch_size, 2048), dtype=tf.float32)

# Create a mock dataset
def create_mock_dataset(num_batches=3, batch_size=20, img_height=4, img_width=3):
    images = []
    for _ in range(num_batches):
        # Each batch: (batch_size, height, width, channels)
        batch = tf.random.normal((batch_size, img_height, img_width, 3), dtype=tf.float32)
        images.append(batch)
    
    # Create a tf.data.Dataset with cardinality
    dataset = tf.data.Dataset.from_tensor_slices(images)
    dataset = dataset.batch(1)  # Each element is one batch
    return dataset

# Build the arguments
mock_inception = MockInception()
mock_dataset = create_mock_dataset(num_batches=3, batch_size=20)

ARGS = {
    "images": mock_dataset,
    "model": mock_inception,
    "batch_size": 20
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
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
calculate_frechet_distance Ran
ran on a synthesized input
milmor/diffusion-transformer-keras/fid.py
code served (permissive licence) · get_code("ddaab4c77b3c0f3a")
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 float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 numpy as np
import warnings

# Set random seed for reproducibility
np.random.seed(0)

# Create small mean vectors (1D arrays)
mu1 = np.array([0.1, 0.2, 0.3, 0.4])
mu2 = np.array([0.15, 0.25, 0.35, 0.45])

# Create small covariance matrices (2D arrays, 4x4)
# Make them positive semi-definite by using outer product with itself
base_matrix_1 = np.random.randn(4, 3)
sigma1 = base_matrix_1 @ base_matrix_1.T + np.eye(4) * 0.1

base_matrix_2 = np.random.randn(4, 3)
sigma2 = base_matrix_2 @ base_matrix_2.T + np.eye(4) * 0.1

# Small epsilon value for numerical stability
eps = 1e-6

ARGS = {
    'mu1': mu1,
    'sigma1': sigma1,
    'mu2': mu2,
    'sigma2': sigma2,
    'eps': eps
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
center_crop_arr Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from liusulin/DDPD
pointer only · get_code("1712a07966b542ee")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

convBlock Ran
ran on a synthesized input
milmor/diffusion-transformer-keras/discriminator.py
code served (permissive licence) · get_code("03149bec2314ebce")
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 Sequential. The image it ran in was not recorded. If your copy returns the same type 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 tensorflow as tf
from tensorflow.keras import layers

ARGS = {
    'filters': 32,
    'kernel_size': 3,
    'initializer': 'glorot_uniform'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
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
create_loader Ran
ran on a synthesized input
milmor/diffusion-transformer/image_datasets.py
code served (permissive licence) · get_code("deedff7c289ab393")
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 _MultiProcessingDataLoaderIter. The image it ran in was not recorded. If your copy returns the same type 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 tempfile
import os
from PIL import Image
import numpy as np

torch.manual_seed(0)
np.random.seed(0)

# Create a temporary directory with test images
temp_dir = tempfile.mkdtemp()

# Create a few small test images
for i in range(3):
    img = Image.new('RGB', (8, 8), color=(i*50, i*80, i*100))
    img.save(os.path.join(temp_dir, f'test_image_{i}.jpg'))

ARGS = {
    'data_dir': temp_dir,
    'img_size': 4,
    'batch_size': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type _MultiProcessingDataLoaderIter — a _MultiProcessingDataLoaderIter output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a _MultiProcessingDataLoaderIter output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
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
deprocess Ran
ran on a synthesized input
milmor/diffusion-transformer/utils.py
code served (permissive licence) · get_code("00999b7ce52789e2")
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 2, dtype family float, shape (2, 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

torch.manual_seed(0)

# The deprocess function takes an img parameter and performs:
# img * 127.5 + 127.5
# This is a simple arithmetic operation on a tensor, so img should be a tensor

img = torch.randn(2, 4)

ARGS = {
    "img": img
}
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 2, dtype family float, shape (2, 4) — shape (2, 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
downBlock Ran
ran on a synthesized input
milmor/diffusion-transformer-keras/autoencoder.py
code served (permissive licence) · get_code("177075a7f51b01e3")
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 Sequential. The image it ran in was not recorded. If your copy returns the same type 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 tensorflow as tf
from tensorflow.keras import layers

# Define arguments for downBlock function
ARGS = {
    'filters': 64,
    'kernel_size': 3,
    'initializer': 'glorot_uniform'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
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
get_1d_sincos_pos_embed_from_grid Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hustvl/dig/models_dig.py
code served (permissive licence) · get_code("e5947aba1d10885f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned ndarray, ndim 2, dtype family float, shape (5, 8) 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 numpy as np

# Create a simple 1D position array
# embed_dim must be even (checked by assert in function)
# pos should be a 1D array of positions

embed_dim = 8  # Small even number for embedding dimension
pos = np.array([0.0, 1.0, 2.0, 3.0, 4.0])  # 5 positions

ARGS = {
    'embed_dim': embed_dim,
    'pos': pos
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (5, 8) — shape (5, 8) 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
get_2d_sincos_pos_embed Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from VinAIResearch/DiMSUM
pointer only · get_code("c92c27c924b517e8")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_activations Ran
ran on a synthesized input
milmor/diffusion-transformer-keras/fid.py
code served (permissive licence) · get_code("b3a0c9d9b68b0ba4")
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 ndarray, ndim 2, dtype family float, shape (6, 2048) 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 numpy as np
import tensorflow as tf

# Create a mock inception model that returns (n_samples, 2048) predictions
class MockInception(tf.Module):
    def __call__(self, batch):
        # batch shape: (batch_size, height, width, channels)
        # return shape: (batch_size, 2048)
        batch_size = tf.shape(batch)[0]
        return tf.random.normal((batch_size, 2048), dtype=tf.float32)

# Create a TensorFlow dataset with 3 batches of size 2
def create_dataset():
    # Create dummy image data: (6, 4, 4, 3) - 6 images, 4x4, 3 channels
    images = tf.random.normal((6, 4, 4, 3), dtype=tf.float32)
    
    # Create dataset and batch it
    dataset = tf.data.Dataset.from_tensor_slices(images)
    dataset = dataset.batch(2)
    
    return dataset

dataset = create_dataset()
inception = MockInception()
batch_size = 2

ARGS = {
    "dataset": dataset,
    "inception": inception,
    "batch_size": batch_size
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (6, 2048) — shape (6, 2048) 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
get_scalings Ran
ran on a synthesized input
milmor/diffusion-transformer/diff_utils.py
code served (permissive licence) · get_code("1a5424f761d38323")
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 tuple, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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

torch.manual_seed(0)

# The function takes two parameters: sig and sig_data
# Both are used in arithmetic operations (**, +, /, sqrt)
# so they should be torch.Tensor objects

sig = torch.tensor([[1.0, 2.0, 3.0, 4.0],
                    [5.0, 6.0, 7.0, 8.0]])

sig_data = torch.tensor([[0.5, 1.5, 2.5, 3.5],
                         [4.5, 5.5, 6.5, 7.5]])

ARGS = {
    "sig": sig,
    "sig_data": sig_data
}
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 tuple, ndim 3, dtype family float — a tuple output has no array shape of its own; none is stated
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
get_sigmas_karras Ran
ran on a synthesized input
milmor/diffusion-transformer/diff_utils.py
code served (permissive licence) · get_code("fa7018efe2948b62")
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 1, dtype family float, shape (11,) 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

torch.manual_seed(0)

ARGS = {
    'n': 10,
    'sigma_min': 0.01,
    'sigma_max': 80.0,
    'rho': 7.0,
    'device': 'cpu'
}
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 1, dtype family float, shape (11,) — shape (11,) 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
modulate Ran
ran; our contract draft was wrong, not the code
milmor/diffusion-transformer/dit.py
code served (permissive licence) · get_code("03310bba324ae4fb")
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) and returned Tensor, ndim 2, dtype family float, shape (2, 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

torch.manual_seed(0)

# Create small tensors for the modulate function
# x: input tensor with shape (batch_size, feature_dim)
x = torch.randn(2, 4)

# shift: tensor that will be unsqueezed to (batch_size, 1)
# so it should have shape (batch_size,)
shift = torch.randn(2)

# scale: tensor that will be unsqueezed to (batch_size, 1)
# so it should have shape (batch_size,)
scale = torch.randn(2)

ARGS = {
    "x": x,
    "shift": shift,
    "scale": scale,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 2, dtype family float, shape (2, 4) — shape (2, 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
shift_scale Ran
ran, honoured the contract we drafted
VachanVY/diffusion-transformer/torch_src/diffusion_transformer.py
code served (permissive licence) · get_code("393c618733caf957")
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) and returned Tensor, ndim 2, dtype family float, shape (2, 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
from torch import nn, Tensor

torch.manual_seed(0)

# Create small tensors for the function
# x: input tensor with shape (batch_size, feature_dim)
x = torch.randn(2, 4)

# gamma: scale tensor with shape (feature_dim,)
gamma = torch.randn(4)

# beta: shift tensor with shape (feature_dim,)
beta = torch.randn(4)

ARGS = {
    "x": x,
    "gamma": gamma,
    "beta": beta,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 2, dtype family float, shape (2, 4) — shape (2, 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
upBlock Ran
ran on a synthesized input
milmor/diffusion-transformer-keras/autoencoder.py
code served (permissive licence) · get_code("dd229bfb0b49b0d5")
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 Sequential. The image it ran in was not recorded. If your copy returns the same type 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 tensorflow as tf
import numpy as np

# Define ARGS dict with all parameters for upBlock
ARGS = {
    'filters': 64,
    'kernel_size': 3,
    'initializer': 'glorot_uniform'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
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
calculate_activation_statistics Not yet run
unverified (harvested, no recorded run)
milmor/diffusion-transformer/fid.py
code served (permissive licence) · get_code("c5a04d1cab5f16a8")
How it ran

no invocation recorded.

create_logger Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from quandao10/slct
pointer only; the stored copy get_code serves, from quandao10/slct, is pointer only (licence: NONE) · get_code("8382f7e48cc8a309")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

get_2d_sincos_pos_embed_from_grid Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from VinAIResearch/DiMSUM
pointer only · get_code("665d8a4e8f673a4c")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

get_activations Not yet run
unverified (harvested, no recorded run)
milmor/diffusion-transformer/fid.py
code served (permissive licence) · get_code("63749a9f8d05d87c")
How it ran

no invocation recorded.

get_diffution_transformer Not yet run
unverified (harvested, no recorded run)
VachanVY/diffusion-transformer/torch_src/diffusion_transformer.py
code served (permissive licence) · get_code("91f16882dada16e4")
How it ran

no invocation recorded.

Also present in other papers' code

Of 20 samples lifted from this paper's linked repositories, 20 carry a function identity and 0 do not. Of the 20 distinct functions with an identity, 8 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 1 the same month, 6 later) and 12 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.

FunctionAlso present inEarliest-datedOthers named
get_2d_sincos_pos_embed_from_grid 190 other papers
17 dated earlier, 2 the same month, 171 later · 65 only as an assembly brick
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
2105.12723 (2021-05)
Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understanding
2111.06377 (2021-11)
Masked Autoencoders Are Scalable Vision Learners
and 187 more not named here
modulate 101 other papers
2 dated earlier, 98 later, 1 undated · 29 only as an assembly brick
2202.04200 (2022-02)
MaskGIT: Masked Generative Image Transformer
2203.10897 (2022-03)
Unified Multivariate Gaussian Mixture for Efficient Neural Image Compression
2303.14389 (2023-03)
Masked Diffusion Transformer is a Strong Image Synthesizer
and 98 more not named here
get_2d_sincos_pos_embed 69 other papers
1 dated earlier, 68 later · 12 only as an assembly brick
2203.10897 (2022-03)
Unified Multivariate Gaussian Mixture for Efficient Neural Image Compression
2303.14389 (2023-03)
Masked Diffusion Transformer is a Strong Image Synthesizer
2305.13311 (2023-05)
VDT: General-purpose Video Diffusion Transformers via Mask Modeling
and 66 more not named here
get_1d_sincos_pos_embed_from_grid 42 other papers
42 later · 22 only as an assembly brick
2303.14389 (2023-03)
Masked Diffusion Transformer is a Strong Image Synthesizer
2305.13311 (2023-05)
VDT: General-purpose Video Diffusion Transformers via Mask Modeling
2310.00426 (2023-10)
PixArt-$α$: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis
and 39 more not named here
center_crop_arr 37 other papers
1 dated earlier, 36 later
2211.17091 (2022-11)
Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models
2311.09257 (2023-11)
UFOGen: You Forward Once Large Scale Text-to-Image Generation via Diffusion GANs
2311.15529 (2023-11)
Efficient Dataset Distillation via Minimax Diffusion
and 34 more not named here
create_logger 14 other papers
14 later
2405.20337 (2024-05)
OccSora: 4D Occupancy Generation Models as World Simulators for Autonomous Driving
2406.01733 (2024-06)
Learning-to-Cache: Accelerating Diffusion Transformer via Layer Caching
2407.11633 (2024-07)
Scaling Diffusion Transformers to 16 Billion Parameters
and 11 more not named here
calculate_activation_statistics 1 other paper
1 dated earlier
2006.10738 (2020-06)
Differentiable Augmentation for Data-Efficient GAN Training
get_activations 1 other paper
1 dated earlier
2006.10738 (2020-06)
Differentiable Augmentation for Data-Efficient GAN Training

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

Figure 1. Diffusion models with transformer backbones achieve state-of-the-art image quality. We show selected samples from two of our class-conditional DiT-XL/2 models trained on ImageNet at 512×512 and 256×256 resolution, respectively.

For agents

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

get_harvested_code_for_paper("2212.09748")
get_code_for_paper("2212.09748")
have("2212.09748")

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

curl https://syntology.ai/api/ran/2212.09748.json

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

[![Syntology run record](https://syntology.ai/api/ran/2212.09748.svg)](https://syntology.ai/paper/2212.09748)

Connect an agent — have() is free.