Papers › Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity

Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity

11 Jan 2021arXiv:2101.03961archive 2025-07-28

William Fedus, Barret Zoph, Noam Shazeer

In deep learning, models typically reuse the same parameters for all inputs. Mixture of Experts (MoE) defies this and instead selects different parameters for each incoming example. The result is a sparsely-activated model -- with outrageous numbers of parameters -- but a constant computational cost. However, despite several notable successes of MoE, widespread adoption has been hindered by complexity, communication costs and training instability -- we address these with the Switch Transformer. We simplify the MoE routing algorithm and design intuitive improved models with reduced communication and computational costs. Our proposed training techniques help wrangle the instabilities and we show large sparse models may be trained, for the first time, with lower precision (bfloat16) formats. We design models based off T5-Base and T5-Large to obtain up to 7x increases in pre-training speed with the same computational resources. These improvements extend into multilingual settings where we measure gains over the mT5-Base version across all 101 languages. Finally, we advance the current scale of language models by pre-training up to trillion parameter models on the "Colossal Clean Crawled Corpus" and achieve a 4x speedup over the T5-XXL model.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 13 code samples: 12 from 3 repositories linked to this paper, 1 identical to code harvested from another paper's repository. Across all 13: 7 ran (of which 2 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 4 where Syntology's instrument failed) · 6 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 4 samples from 1 repository, 0 ran (0 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 8 samples from 2 repositories, 6 ran (3 with no instrument failure; 3 where Syntology's instrument failed); 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 3 of its repositories that Syntology harvested for other papers:

Levels written on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 7 samples that ran has no recorded date.

tensorflow/mesh officialmentioned in papertfSyntology: 4 samples, 0 ran · 4 unverified report
efeslab/fiddler mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
lingjzhu/charsiug2p mentioned on GitHubpytorchSyntology: not harvested report
vikparuchuri/marker mentioned on GitHubpytorchSyntology: not harvested report
zyphra/blackmamba mentioned on GitHubpytorchSyntology: 7 samples, 5 ran (3 instrument-failed) · 2 unverified report
MindCode-4/code-5 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ViTamin: Designing Scalable Vision Models in the Vision-Language Era (arXiv:2404.02132; that paper's own run record: 6 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 1 violated, 3 with no contract checked; 2 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method. report
labmlai/annotated_deep_learning_paper_implementations pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Playing Atari with Deep Reinforcement Learning (arXiv:1312.5602; that paper's own run record: 64 ran (of which 24 constructed an object rather than computing a result; 46 with no instrument failure: 5 honoured, 0 violated, 41 with no contract checked; 18 where Syntology's instrument failed) · 53 unverified), Adam: A Method for Stochastic Optimization (arXiv:1412.6980; that paper's own run record: 27 ran (of which 7 constructed an object rather than computing a result; 23 with no instrument failure: 4 honoured, 2 violated, 17 with no contract checked; 4 where Syntology's instrument failed) · 10 unverified), Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift (arXiv:1502.03167; that paper's own run record: 19 ran (of which 5 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified) and 12 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
pwc-1/Paper-10 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large... (arXiv:2403.07714; that paper's own run record: 11 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 2 violated, 6 with no contract checked; 3 where Syntology's instrument failed) · 4 unverified); nothing checks that they implement this paper's method. report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

13 samples harvested: 7 ran (of which 2 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 4 where Syntology's instrument failed) · 6 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

4ran · our draft was wrong
3ran
6unverified

Licence: 8 of the 13 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 3 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. 1 sample is listed last, in a group without a repository. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 1 sample that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

tensorflow/mesh official repository 4 samples, 0 ran · 4 unverified
_add_token_emb_to_gate_inputs mesh_tensorflow/transformer/moe.py unverified Apache-2.0 (permissive) · 898627b88bfd6d44 · report
_router_z_loss mesh_tensorflow/transformer/moe.py unverified Apache-2.0 (permissive) · 7179ac118906b2a5 · report
_stochastically_use_non_top_expert mesh_tensorflow/transformer/moe.py unverified Apache-2.0 (permissive) · da4ea018ee62163d · report
_switch_gating mesh_tensorflow/transformer/moe.py unverified Apache-2.0 (permissive) · 2d4e03e49ac8f264 · report
efeslab/fiddler community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
MixtralBlockSparseTop2MLP benchmarks/mixtral_offloading/src/custom_layers.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · 74374c911af39f6c · report fixture: signature hidden_states:Tensor/3/float · returned Tensor · level recorded 2026-09-22
zyphra/blackmamba community (archive-listed) 7 samples, 5 ran (3 instrument-failed) · 2 unverified
MLP in 375 papers switch_mlp.py ran no licence file found · pointer only · ff4dfeb7df35fff6 · report fixture: returned MLP · level recorded 2026-09-22
MambaConfig switch_mlp.py ran no licence file found · pointer only · 1db09a0131e40faf · report fixture: returned MambaConfig · level recorded 2026-09-22
init_method_normal in 7 papers switch_mlp.py ran · our draft was wrong no licence file found · pointer only · 58a3d812be2f2018 · report fixture: signature sigma:float/0/float · returned function · level recorded 2026-09-22
scaled_init_method_normal switch_mlp.py ran · our draft was wrong no licence file found · pointer only · c319770f5ead4490 · report fixture: signature sigma:float/0/float, num_layers:int/0/int · returned function · level recorded 2026-09-22
sinkhorn in 47 papers
census: 11 implementations from 11 papers, 11 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 2, kind float, dtype float32
switch_mlp.py ran · our draft was wrong fingerprinted no licence file found · pointer only · bfbd162761e2e475 · report fixture: signature cost:Tensor/2/float, tol:float/0/float · returned Tensor · level recorded 2026-09-22
SwitchMLP switch_mlp.py unverified no licence file found · pointer only · 02e5457338807a13 · report
selective_scan_fn in 6 papers ops/selective_scan_interface.py unverified no licence file found · pointer only · c48b9d686b548496 · report
Repository not located 1 sample, 1 ran (1 instrument-failed) · 0 unverified
1 sample is identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
selective_scan_ref in 14 papers repository not located ran · our draft was wrong fingerprinted licence of this copy not recorded · ba7b868fec181b57 · report fixture: signature u:Tensor/3/float, delta:Tensor/3/float, A:Tensor/2/float, B:Tensor/3/float, C:Tensor/3/float, D:Tensor/1/float, z:Tensor/3/float, delta_bias:Tensor/1/float, delta_softplus:bool/0/bool, return_last_state:bool/0/bool · returned Tensor · level recorded, date not recorded

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

3 assemblies on this paper: 3 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

3 sibling assemblies for Switch Transformer (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Switch Transformer (graph method: Switch Transformer). 6 bricks; individually on a fixture: 5 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. SwitchMLP entry · level 0 · unverified · 02e5457338807a13
  2. MLP helper · level 1 · ran · ff4dfeb7df35fff6
  3. MambaConfig helper · level 1 · ran · 1db09a0131e40faf
  4. sinkhorn helper · level 1 · ran · our draft was wrong · bfbd162761e2e475
  5. init_method_normal helper · level 1 · ran · our draft was wrong · 58a3d812be2f2018
  6. scaled_init_method_normal helper · level 1 · ran · our draft was wrong · c319770f5ead4490

harvested Assembly for Switch Transformer (graph method: Switch Transformer). 4 bricks; individually on a fixture: 0 ran · 4 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. _switch_gating entry · level 0 · unverified · 2d4e03e49ac8f264
  2. _add_token_emb_to_gate_inputs helper · level 0 · unverified · 898627b88bfd6d44
  3. _router_z_loss loss · level 0 · unverified · 7179ac118906b2a5
  4. _stochastically_use_non_top_expert helper · level 0 · unverified · da4ea018ee62163d

harvested Assembly for Switch Transformer (graph method: Switch Transformer). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. MixtralBlockSparseTop2MLP entry · level 1 · ran · 74374c911af39f6c

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2101.03961") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Language ModellingMixture-of-ExpertsQuestion Answering

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Introduced by this paper: Switch FFN

Switch FFNSwitch Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections