SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1602.07868 · 2016

Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks

arXiv · PDF · Open in the Atlas

Code that ran

10 of 16 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 9 have no contract verdict; 1 failure of our instrument; 6 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/1602.07868.json · badge: /api/ran/1602.07868.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
openai/weightnorm canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
krasserm/wdsr community (classifier label)
classifier basis not recorded on the link
6 ran · 2 no recorded run
UrosOgrizovic/SimpleGoogleQuickdraw community (classifier label)
classifier basis not recorded on the link
3 ran · 3 no recorded run
zoli333/Weight-Normalization reimplementation (classifier label)
classifier basis: classifier_2025
0 ran · 1 no recorded run
FunctionStatusWhere it lives
get_image_from_images Ran
ran on a synthesized input
UrosOgrizovic/SimpleGoogleQuickdraw/image_operations.py
code served (permissive licence) · get_code("c84c1ed8059aa2ed")
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 (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 function simply indexes into 'images' using 'index'
# So 'images' needs to be indexable (list, tensor, etc.)
# and 'index' needs to be a valid index

# Create a small tensor as the images container
images = torch.randn(2, 4)  # batch size 2, dimension 4

# Create a valid index
index = 0

ARGS = {
    "images": images,
    "index": index
}
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 (4,) — shape (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
int_shape Ran
ran; our contract draft was wrong, not the code
openai/weightnorm/tensorflow/nn.py
code served (permissive licence) · get_code("3ba6aec873833b39")
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 list, ndim 1, dtype family int; 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
import tensorflow as tf

torch.manual_seed(0)

# The function int_shape(x) calls x.get_shape() and converts to list of ints
# This is a TensorFlow pattern, so we need a TensorFlow tensor or object with get_shape()

# Create a TensorFlow tensor with a small shape
x = tf.constant([[1.0, 2.0, 3.0, 4.0],
                 [5.0, 6.0, 7.0, 8.0]])

ARGS = {
    'x': x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
tensorflow, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list 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
load_image Ran
ran on a synthesized input
krasserm/wdsr/utils.py
code served (permissive licence) · get_code("0ab4dee93051e84f")
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 3, dtype family int, shape (4, 3, 3) 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
from PIL import Image
import tempfile
import os

# Create a temporary image file
temp_dir = tempfile.mkdtemp()
temp_image_path = os.path.join(temp_dir, "test_image.png")

# Create a small test image and save it
test_image_array = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)
test_image = Image.fromarray(test_image_array)
test_image.save(temp_image_path)

ARGS = {
    "path": temp_image_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (4, 3, 3) — shape (4, 3, 3) 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
load_images Ran
ran on a synthesized input
UrosOgrizovic/SimpleGoogleQuickdraw/image_operations.py
code served (permissive licence) · get_code("2d732893414e2966")
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 list, ndim 3, dtype family int; 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 tempfile
import os

# Create a temporary numpy file with fake image data
# Images should be flattened 28x28 = 784 values each
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_images.npy")

# Create a small array of fake images (5 images, each 784 values)
fake_images = np.random.randint(0, 256, size=(5, 784)).astype(np.uint8)
np.save(temp_file, fake_images)

ARGS = {
    "file_path": temp_file,
    "how_many": 3,
    "is_transfer_learning": False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 3, dtype family int — a list 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
pad_image Ran
ran on a synthesized input
UrosOgrizovic/SimpleGoogleQuickdraw/image_operations.py
code served (permissive licence) · get_code("ffc86641164a4445")
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 (32, 32) 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

# The function pad_image takes an img parameter
# It reshapes img to (28, 28), then pads it with 2 pixels on all sides
# So img needs to be a numpy array (or array-like) with at least 784 elements

# Create a small 28x28 image as a flattened array
img_data = np.random.rand(784)

ARGS = {
    "img": img_data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 (32, 32) — shape (32, 32) 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
random_crop Ran
ran on a synthesized input
krasserm/wdsr/data.py
code served (permissive licence) · get_code("40c578a7c8a918e6")
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 tensorflow as tf
import numpy as np

# Create small test images
# LR image: 100x100 (low resolution)
# HR image: 200x200 (high resolution, 2x upscaled)
lr_img = tf.constant(np.random.randint(0, 256, (100, 100, 3), dtype=np.uint8), dtype=tf.uint8)
hr_img = tf.constant(np.random.randint(0, 256, (200, 200, 3), dtype=np.uint8), dtype=tf.uint8)

ARGS = {
    "lr_img": lr_img,
    "hr_img": hr_img,
    "hr_crop_size": 96,
    "scale": 2,
}
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
random_flip Ran
ran on a synthesized input
krasserm/wdsr/data.py
code served (permissive licence) · get_code("256bed53fab6520f")
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 tensorflow as tf

# Create small test images for lr_img and hr_img
# lr_img: low-resolution image, shape (batch, height, width, channels)
# hr_img: high-resolution image, shape (batch, height, width, channels)
lr_img = tf.constant([[[[1.0, 2.0, 3.0],
                        [4.0, 5.0, 6.0],
                        [7.0, 8.0, 9.0],
                        [10.0, 11.0, 12.0]],
                       [[13.0, 14.0, 15.0],
                        [16.0, 17.0, 18.0],
                        [19.0, 20.0, 21.0],
                        [22.0, 23.0, 24.0]]]], dtype=tf.float32)

hr_img = tf.constant([[[[1.0, 2.0],
                        [3.0, 4.0]],
                       [[5.0, 6.0],
                        [7.0, 8.0]]]], dtype=tf.float32)

ARGS = {
    'lr_img': lr_img,
    'hr_img': hr_img
}
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 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
random_rotate Ran
ran on a synthesized input
krasserm/wdsr/data.py
code served (permissive licence) · get_code("4cb58354ae7610d2")
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 tensorflow as tf
import numpy as np

# Create small sample images for low-resolution and high-resolution
# lr_img: low-resolution image, shape (height, width, channels)
# hr_img: high-resolution image, shape (height, width, channels)
lr_img = tf.constant(np.random.rand(4, 3, 3).astype(np.float32))
hr_img = tf.constant(np.random.rand(8, 6, 3).astype(np.float32))

ARGS = {
    "lr_img": lr_img,
    "hr_img": hr_img,
}
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
resolve Ran
ran on a synthesized input
krasserm/wdsr/model/common.py
code served (permissive licence) · get_code("fcb21bb87592664b")
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 EagerTensor, ndim 4, dtype family int, shape (2, 4, 3, 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 tensorflow as tf
import numpy as np

# Create a minimal model that takes a low-resolution batch and outputs super-resolved images
class SimpleSuperResolutionModel(tf.Module):
    def __init__(self):
        super().__init__()
        self.dense = tf.keras.layers.Dense(12, activation='relu')
        self.out = tf.keras.layers.Dense(12)
    
    def __call__(self, x):
        # x shape: (batch, height, width, channels)
        # Flatten, process, and reshape back
        batch_size = tf.shape(x)[0]
        original_shape = tf.shape(x)
        
        x_flat = tf.reshape(x, [batch_size, -1])
        x = self.dense(x_flat)
        x = self.out(x)
        
        # Reshape to match expected output shape (same as input for simplicity)
        sr = tf.reshape(x, original_shape)
        return sr

# Create the model
model = SimpleSuperResolutionModel()

# Create a small low-resolution batch
# Shape: (batch_size=2, height=4, width=3, channels=1)
# Using float32 as the function casts to float32 anyway
lr_batch = tf.constant(
    np.random.randint(0, 100, size=(2, 4, 3, 1), dtype=np.int32),
    dtype=tf.int32
)

ARGS = {
    "model": model,
    "lr_batch": lr_batch
}
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 EagerTensor, ndim 4, dtype family int, shape (2, 4, 3, 1) — shape (2, 4, 3, 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
resolve_single Ran
ran on a synthesized input
krasserm/wdsr/model/common.py
code served (permissive licence) · get_code("ae235fb465d0b309")
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 EagerTensor, ndim 3, dtype family int, shape (2, 2, 3) 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 tensorflow as tf
import numpy as np

# Create a minimal model that takes low-res input and outputs super-resolved output
class SimpleUpscaleModel(tf.Module):
    def __init__(self):
        super(SimpleUpscaleModel, self).__init__()
        self.dense = tf.keras.layers.Dense(12, activation='relu')
        self.output_layer = tf.keras.layers.Dense(12)
    
    def __call__(self, lr_batch):
        # lr_batch shape: (batch, height, width, channels) -> (batch, 12)
        x = tf.reshape(lr_batch, [tf.shape(lr_batch)[0], -1])
        x = self.dense(x)
        # Output should be in range [0, 255] after clipping and rounding
        x = self.output_layer(x)
        # Reshape back to image-like format
        x = tf.reshape(x, [tf.shape(lr_batch)[0], 2, 2, 3])
        return x

# Instantiate the model
model = SimpleUpscaleModel()

# Create a low-res input image (batch_size=1, height=2, width=2, channels=3)
# Values in uint8 range for realism
lr = tf.constant(np.random.randint(0, 256, (2, 2, 3), dtype=np.uint8), dtype=tf.uint8)

ARGS = {
    "model": model,
    "lr": lr
}
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 EagerTensor, ndim 3, dtype family int, shape (2, 2, 3) — shape (2, 2, 3) 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
data_dependent_weight_norm_g_init Not yet run
unverified (harvested, no recorded run)
zoli333/Weight-Normalization/model.py
pointer only (licence: NONE) · get_code("5be6863211220d86")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zoli333/Weight-Normalization when it was checked. 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.

evaluate Not yet run
unverified (harvested, no recorded run)
krasserm/wdsr/model/common.py
code served (permissive licence) · get_code("e0f9b50298061760")
How it ran

no invocation recorded.

get_model Not yet run
unverified (harvested, no recorded run)
UrosOgrizovic/SimpleGoogleQuickdraw/models/SVM/SVM.py
code served (permissive licence) · get_code("0651dc674036b1ce")
How it ran

no invocation recorded.

get_model Not yet run
unverified (harvested, no recorded run)
UrosOgrizovic/SimpleGoogleQuickdraw/models/transfer_learning/transfer_learning.py
code served (permissive licence) · get_code("6f81146fd4122ca8")
How it ran

no invocation recorded.

get_model Not yet run
unverified (harvested, no recorded run)
UrosOgrizovic/SimpleGoogleQuickdraw/models/vanilla_cnn/vanilla_cnn.py
code served (permissive licence) · get_code("a771fe43d3f305e2")
How it ran

no invocation recorded.

res_block Not yet run
unverified (harvested, no recorded run)
krasserm/wdsr/model/edsr.py
code served (permissive licence) · get_code("71ba60506facd18f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 9 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 9 later) and 7 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
int_shape 4 other papers
4 later
1605.08803 (2016-05)
Density estimation using Real NVP
1702.08658 (2017-02)
Towards Deeper Understanding of Variational Autoencoding Models
1807.10589 (2018-07)
Diverse feature visualizations reveal invariances in early layers of deep neural networks
and 1 more not named here
evaluate 1 other paper
1 later
1808.08718 (2018-08)
Wide Activation for Efficient and Accurate Image Super-Resolution
load_image 1 other paper
1 later
1808.08718 (2018-08)
Wide Activation for Efficient and Accurate Image Super-Resolution
random_crop 1 other paper
1 later
1808.08718 (2018-08)
Wide Activation for Efficient and Accurate Image Super-Resolution
random_flip 1 other paper
1 later
1808.08718 (2018-08)
Wide Activation for Efficient and Accurate Image Super-Resolution
random_rotate 1 other paper
1 later
1808.08718 (2018-08)
Wide Activation for Efficient and Accurate Image Super-Resolution
res_block 1 other paper
1 later
1808.08718 (2018-08)
Wide Activation for Efficient and Accurate Image Super-Resolution
resolve 1 other paper
1 later
1808.08718 (2018-08)
Wide Activation for Efficient and Accurate Image Super-Resolution
resolve_single 1 other paper
1 later
1808.08718 (2018-08)
Wide Activation for Efficient and Accurate Image Super-Resolution

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

We present weight normalization: a reparameterization of the weight vectors in a neural network that decouples the length of those weight vectors from their direction. By reparameterizing the weights in this way we improve the conditioning of the optimization problem and we speed up convergence of stochastic gradient descent. Our reparameterization is inspired by batch normalization but does not introduce any dependencies between the examples in a minibatch. This means that our method can also be applied successfully to recurrent models such as LSTMs and to noise-sensitive applications such as deep reinforcement learning or generative models, for which batch normalization is less well suited. Although our method is much simpler, it still provides much of the speed-up of full batch normalization. In addition, the computational overhead of our method is lower, permitting more optimization steps to be taken in the same amount of time. We demonstrate the usefulness of our method on applications in supervised image recognition, generative modelling, and deep reinforcement learning.

For agents

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

get_harvested_code_for_paper("1602.07868")
get_code_for_paper("1602.07868")
have("1602.07868")

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

curl https://syntology.ai/api/ran/1602.07868.json

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

[![Syntology run record](https://syntology.ai/api/ran/1602.07868.svg)](https://syntology.ai/paper/1602.07868)

Connect an agent — have() is free.