SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2412.11549 · 2024

MPQ-DM: Mixed Precision Quantization for Extremely Low Bit Diffusion Models

arXiv · PDF · Open in the Atlas

Code that ran

11 of 13 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, 4 violated it and 0 have no contract verdict; 7 failures of our instrument; 2 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2412.11549.json · badge: /api/ran/2412.11549.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
cantbebetter2/mpq-dm canonical (classifier label)
classifier basis: census
11 ran · 2 no recorded run
FunctionStatusWhere it lives
round_ste Ran
ran; our contract draft was wrong, not the code
cantbebetter2/mpq-dm/quant_scripts/quant_layer.py
code served (permissive licence) · get_code("675da8641c602b03")
How it ran

Ran in Syntology's jail 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. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-05T00:12:31Z). 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 takes a single parameter: x, which is a torch.Tensor
# It performs rounding with straight-through estimator
# We need a tensor with some values that will be rounded

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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
always Ran
ran; our contract draft was wrong, not the code
cantbebetter2/mpq-dm/ldm/modules/x_transformer.py
code served (permissive licence) · get_code("fe5dd5258046898c")
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 function. 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.

ARGS = {
    'val': 42
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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_kurtosis_per_channel Ran
ran; our contract draft was wrong, not the code
cantbebetter2/mpq-dm/quant_scripts/quant_layer.py
code served (permissive licence) · get_code("8ff42a5748d7cb96")
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)

# Create a small tensor for data
# Shape: (batch_size=2, channels=4, sequence_length=3)
# This gives us a 3D tensor where we compute kurtosis per channel
data = torch.randn(2, 4, 3)

ARGS = {
    "data": 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 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
default Ran
ran, violated the contract we drafted
cantbebetter2/mpq-dm/ldm/modules/attention.py
code served (permissive licence) · get_code("424012cb37b31172")
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 int, ndim 0, dtype family int, shape () 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.

from inspect import isfunction

def exists(val):
    return val is not None

def default(val, d):
    if exists(val):
        return val
    return d() if isfunction(d) else d

# Test case 1: val exists (not None), d is a regular value
ARGS_1 = {
    'val': 42,
    'd': 100
}

# Test case 2: val is None, d is a regular value
ARGS_2 = {
    'val': None,
    'd': 100
}

# Test case 3: val is None, d is a function
def default_func():
    return 200

ARGS_3 = {
    'val': None,
    'd': default_func
}

# Test case 4: val exists, d is a function
ARGS_4 = {
    'val': "existing_value",
    'd': lambda: "default_value"
}

# Use one of the test cases as the final ARGS
ARGS = ARGS_3
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, shape () — shape () 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
exists Ran
ran, violated the contract we drafted
cantbebetter2/mpq-dm/ldm/modules/attention.py
pointer only (licence: MIT); the stored copy get_code serves, from ThomasMrY/VCT, is pointer only (licence: NONE) · get_code("aa5486a3650902d8")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ThomasMrY/VCT, and GitHub detected no licence in that repository 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. A run is recorded.

get_train_samples Ran
ran; our contract draft was wrong, not the code
cantbebetter2/mpq-dm/quant_scripts/sample_lora_model.py
pointer only (licence: MIT); the stored copy get_code serves, from ziplab/PTQD, is pointer only (licence: NONE) · get_code("8dc011d26607845d")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ziplab/PTQD, and GitHub detected no licence in that repository 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. A run is recorded.

get_train_samples Ran
ran; our contract draft was wrong, not the code
cantbebetter2/mpq-dm/quant_scripts/sample_lora_model_lsun.py
code served (permissive licence) · get_code("a17aa236f371840f")
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 torch

torch.manual_seed(0)

# Create a simple data loader as a list of (image, t) tuples
# The function iterates over train_loader and unpacks each element as (image, t)
# It then concatenates along dim=0, so we need batch dimension first

train_loader = [
    (torch.randn(2, 4), torch.randn(2, 3)),  # batch_size=2, image_dim=4, t_dim=3
    (torch.randn(2, 4), torch.randn(2, 3)),  # batch_size=2, image_dim=4, t_dim=3
    (torch.randn(2, 4), torch.randn(2, 3)),  # batch_size=2, image_dim=4, t_dim=3
]

num_samples = 3

ARGS = {
    "train_loader": train_loader,
    "num_samples": num_samples,
}
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 — 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
isimage Ran
ran, violated the contract we drafted
cantbebetter2/mpq-dm/ldm/util.py
code served (permissive licence) · get_code("b1368330cf0f5642")
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 bool, ndim 0, dtype family float, shape () 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 isimage checks if x is a torch.Tensor with 4 dimensions
# and the second dimension (channel dim) is either 3 or 1
# Test case: valid image tensor with shape (batch, channels, height, width)
x = torch.randn(2, 3, 4, 5)

ARGS = {
    "x": x
}
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 bool, ndim 0, dtype family float, shape () — shape () 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
ismap Ran
ran, violated the contract we drafted
cantbebetter2/mpq-dm/ldm/util.py
code served (permissive licence) · get_code("d72762b700feee6f")
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 bool, ndim 0, dtype family float, shape () 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 checks if x is a torch.Tensor with 4 dimensions where the second dimension (x.shape[1]) is > 3
# We need to construct a tensor that satisfies this condition

# Create a tensor with shape (batch=2, channels=5, height=4, width=3)
# This has 4 dimensions and x.shape[1] = 5 > 3, so it should return True
x = torch.randn(2, 5, 4, 3)

ARGS = {
    'x': x
}
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 bool, ndim 0, dtype family float, shape () — shape () 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
lp_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cantbebetter2/mpq-dm/quant_scripts/quant_layer.py
code served (permissive licence) · get_code("08c6ea7576220a13")
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 0, dtype family float, shape () 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)

# Create small tensors for pred and tgt
# pred: predictions, shape (batch_size, features)
# tgt: targets, shape (batch_size, features)
batch_size = 2
feature_dim = 4

pred = torch.randn(batch_size, feature_dim)
tgt = torch.randn(batch_size, feature_dim)

# Parameters for the loss function
p = 2.0
reduction = 'none'

ARGS = {
    'pred': pred,
    'tgt': tgt,
    'p': p,
    'reduction': reduction
}
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 0, dtype family float, shape () — shape () 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
uniq Ran
ran; our contract draft was wrong, not the code
cantbebetter2/mpq-dm/ldm/modules/attention.py
pointer only (licence: MIT); the stored copy get_code serves, from JingWu321/MUNBa, is pointer only (licence: NONE) · get_code("9a299fe5ae09e407")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from JingWu321/MUNBa, and GitHub detected no licence in that repository 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. A run is recorded.

getStep Not yet run
unverified (harvested, no recorded run)
cantbebetter2/mpq-dm/ldm/globalvar.py
code served (permissive licence) · get_code("80ae154054c233d5")
How it ran

no invocation recorded.

log_txt_as_img Not yet run
unverified (harvested, no recorded run)
cantbebetter2/mpq-dm/ldm/util.py
code served (permissive licence) · get_code("f9bd2e83191afad1")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 11 are also present in other papers whose code we hold (11 with at least one paper dated earlier, 7 the same month, 9 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.

FunctionAlso present inEarliest-datedOthers named
exists 486 other papers
412 dated earlier, 10 the same month, 53 later, 9 same year, order not determinable, 2 undated · 78 only as an assembly brick
1603.08155 (2016-03)
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
1611.07004 (2016-11)
Image-to-Image Translation with Conditional Adversarial Networks
1804.04637 (2018-04)
EMBER: An Open Dataset for Training Static PE Malware Machine Learning Models
and 483 more not named here
default 207 other papers
177 dated earlier, 4 the same month, 20 later, 6 same year, order not determinable · 41 only as an assembly brick
1711.11585 (2017-11)
High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.08162 (2020-10)
SidechainNet: An All-Atom Protein Structure Dataset for Machine Learning
and 204 more not named here
uniq 115 other papers
101 dated earlier, 4 the same month, 6 later, 4 same year, order not determinable
2112.10752 (2021-12)
High-Resolution Image Synthesis with Latent Diffusion Models
2204.06125 (2022-04)
Hierarchical Text-Conditional Image Generation with CLIP Latents
2206.02779 (2022-06)
Blended Latent Diffusion
and 112 more not named here
ismap 69 other papers
60 dated earlier, 3 the same month, 3 later, 3 same year, order not determinable
2204.06125 (2022-04)
Hierarchical Text-Conditional Image Generation with CLIP Latents
2206.02779 (2022-06)
Blended Latent Diffusion
2208.13753 (2022-08)
Frido: Feature Pyramid Diffusion for Complex Scene Image Synthesis
and 66 more not named here
isimage 61 other papers
53 dated earlier, 3 the same month, 3 later, 2 same year, order not determinable
2206.02779 (2022-06)
Blended Latent Diffusion
2208.13753 (2022-08)
Frido: Feature Pyramid Diffusion for Complex Scene Image Synthesis
2210.12965 (2022-10)
High-Resolution Image Editing via Multi-Stage Blended Diffusion
and 58 more not named here
always 56 other papers
51 dated earlier, 2 the same month, 2 later, 1 same year, order not determinable
2102.12092 (2021-02)
Zero-Shot Text-to-Image Generation
2107.03312 (2021-07)
SoundStream: An End-to-End Neural Audio Codec
2206.00100 (2022-06)
VALHALLA: Visual Hallucination for Machine Translation
and 53 more not named here
log_txt_as_img 55 other papers
46 dated earlier, 3 the same month, 3 later, 3 same year, order not determinable
2204.06125 (2022-04)
Hierarchical Text-Conditional Image Generation with CLIP Latents
2206.02779 (2022-06)
Blended Latent Diffusion
2210.12965 (2022-10)
High-Resolution Image Editing via Multi-Stage Blended Diffusion
and 52 more not named here
round_ste 27 other papers
19 dated earlier, 8 later · 13 only as an assembly brick
2203.05740 (2022-03)
QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantization
2204.12322 (2022-04)
RAPQ: Rescuing Accuracy for Power-of-Two Low-bit Post-training Quantization
and 24 more not named here
lp_loss 18 other papers
13 dated earlier, 5 later · 7 only as an assembly brick
Li_RepQ-ViT_Scale_Reparameterization_for_Post-Training_Quantization_of_Vision_Transformers_ICCV_2023_paper (2023 (year only))
RepQ-ViT: Scale Reparameterization for Post-Training Quantization of Vision Transformers
2302.04304 (2023-02)
Q-Diffusion: Quantizing Diffusion Models
and 15 more not named here
get_train_samples 2 other papers
2 dated earlier
2305.10657 (2023-05)
PTQD: Accurate Post-Training Quantization for Diffusion Models
2310.03270 (2023-10)
EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models
getStep 1 other paper
1 dated earlier
2310.03270 (2023-10)
EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models

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

Diffusion models have received wide attention in generation tasks. However, the expensive computation cost prevents the application of diffusion models in resource-constrained scenarios. Quantization emerges as a practical solution that significantly saves storage and computation by reducing the bit-width of parameters. However, the existing quantization methods for diffusion models still cause severe degradation in performance, especially under extremely low bit-widths (2-4 bit). The primary decrease in performance comes from the significant discretization of activation values at low bit quantization. Too few activation candidates are unfriendly for outlier significant weight channel quantization, and the discretized features prevent stable learning over different time steps of the diffusion model. This paper presents MPQ-DM, a Mixed-Precision Quantization method for Diffusion Models. The proposed MPQ-DM mainly relies on two techniques:(1) To mitigate the quantization error caused by outlier severe weight channels, we propose an Outlier-Driven Mixed Quantization (OMQ) technique that uses $Kurtosis$ to quantify outlier salient channels and apply optimized intra-layer mixed-precision bit-width allocation to recover accuracy performance within target efficiency.(2) To robustly learn representations crossing time steps, we construct a Time-Smoothed Relation Distillation (TRD) scheme between the quantized diffusion model and its full-precision counterpart, transferring discrete and continuous latent to a unified relation space to reduce the representation inconsistency. Comprehensive experiments demonstrate that MPQ-DM achieves significant accuracy gains under extremely low bit-widths compared with SOTA quantization methods. MPQ-DM achieves a 58\% FID decrease under W2A4 setting compared with baseline, while all other methods even collapse.

For agents

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

get_harvested_code_for_paper("2412.11549")
get_code_for_paper("2412.11549")
have("2412.11549")

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

curl https://syntology.ai/api/ran/2412.11549.json

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

[![Syntology run record](https://syntology.ai/api/ran/2412.11549.svg)](https://syntology.ai/paper/2412.11549)

Connect an agent — have() is free.