SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2304.11277 · 2023

PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel

arXiv · PDF · Open in the Atlas

Code that ran

2 of 7 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 1 has no contract verdict; 1 failure 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/2304.11277.json · badge: /api/ran/2304.11277.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
pytorch-labs/applied-ai community (classifier label)
classifier basis not recorded on the link
2 ran · 2 no recorded run
copy not recorded community
not in the paper's declared repositories
0 ran · 3 no recorded run
FunctionStatusWhere it lives
bf16_bench Ran
ran on a synthesized input
pytorch-labs/applied-ai/kernels/needs_perf_help/fp8_gemm_bench.py
code served (permissive licence) · get_code("5583f70605dd1824")
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 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.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create small tensors for matrix multiplication
# x: (batch, input_dim) = (2, 4)
# w: (output_dim, input_dim) = (3, 4)
# matmul(x, w.T) will produce (2, 3)

x = torch.randn(2, 4, dtype=torch.bfloat16)
w = torch.randn(3, 4, dtype=torch.bfloat16)

ARGS = {
    "x": x,
    "w": w,
}
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 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
init_to_zero Ran
ran; our contract draft was wrong, not the code
pytorch-labs/applied-ai/kernels/needs_perf_help/fp8_rowwise_tma_persistent.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from amazon-science/dualkv-flash-attn-for-rl, is pointer only (licence: NOASSERTION) · get_code("1586b639a5abd0b0")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from amazon-science/dualkv-flash-attn-for-rl, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

benchmark_sm100_vs_torch Not yet run
unverified (harvested, no recorded run)
pytorch-labs/applied-ai/kernels/blackwell/cute_gemm_01/driver.py
code served (permissive licence) · get_code("0950e350f7edf923")
How it ran

no invocation recorded.

register_module_buffer_registration_hook Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from pytorch/pytorch
pointer only; the stored copy get_code serves, from pytorch/pytorch, is pointer only (licence: NOASSERTION) · get_code("358905d8834f7cfa")
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.

register_module_module_registration_hook Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from pytorch/pytorch
pointer only; the stored copy get_code serves, from pytorch/pytorch, is pointer only (licence: NOASSERTION) · get_code("2e8b412bf625c913")
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.

register_module_parameter_registration_hook Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from pytorch/pytorch
pointer only; the stored copy get_code serves, from pytorch/pytorch, is pointer only (licence: NOASSERTION) · get_code("0a1bb6fa560c4b8d")
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.

sm100_gemm_f16 Not yet run
unverified (harvested, no recorded run)
pytorch-labs/applied-ai/kernels/blackwell/cute_gemm_01/driver.py
code served (permissive licence) · get_code("8850dbb5630032ce")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 4 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 1 later) and 3 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
init_to_zero 12 other papers
12 later
2309.06180 (2023-09)
Efficient Memory Management for Large Language Model Serving with PagedAttention
2310.05869 (2023-10)
HyperAttention: Long-context Attention in Near-Linear Time
2312.15685 (2023-12)
What Makes Good Data for Alignment? A Comprehensive Study of Automatic Data Selection in Instruction Tuning
and 9 more not named here
register_module_buffer_registration_hook 2 other papers
2 dated earlier
1912.01703 (2019-12)
PyTorch: An Imperative Style, High-Performance Deep Learning Library
2006.15704 (2020-06)
PyTorch Distributed: Experiences on Accelerating Data Parallel Training
register_module_module_registration_hook 2 other papers
2 dated earlier
1912.01703 (2019-12)
PyTorch: An Imperative Style, High-Performance Deep Learning Library
2006.15704 (2020-06)
PyTorch Distributed: Experiences on Accelerating Data Parallel Training
register_module_parameter_registration_hook 2 other papers
2 dated earlier
1912.01703 (2019-12)
PyTorch: An Imperative Style, High-Performance Deep Learning Library
2006.15704 (2020-06)
PyTorch Distributed: Experiences on Accelerating Data Parallel 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

It is widely acknowledged that large models have the potential to deliver superior performance across a broad range of domains. Despite the remarkable progress made in the field of machine learning systems research, which has enabled the development and exploration of large models, such abilities remain confined to a small group of advanced users and industry leaders, resulting in an implicit technical barrier for the wider community to access and leverage these technologies. In this paper, we introduce PyTorch Fully Sharded Data Parallel (FSDP) as an industry-grade solution for large model training. FSDP has been closely co-designed with several key PyTorch core components including Tensor implementation, dispatcher system, and CUDA memory caching allocator, to provide non-intrusive user experiences and high training efficiency. Additionally, FSDP natively incorporates a range of techniques and settings to optimize resource utilization across a variety of hardware configurations. The experimental results demonstrate that FSDP is capable of achieving comparable performance to Distributed Data Parallel while providing support for significantly larger models with near-linear scalability in terms of TFLOPS.

For agents

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

get_harvested_code_for_paper("2304.11277")
get_code_for_paper("2304.11277")
have("2304.11277")

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

curl https://syntology.ai/api/ran/2304.11277.json

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

[![Syntology run record](https://syntology.ai/api/ran/2304.11277.svg)](https://syntology.ai/paper/2304.11277)

Connect an agent — have() is free.