Papers › Continuous Thought Machines

Continuous Thought Machines

8 May 2025arXiv:2505.05522archive 2025-07-28

Luke Darlow, Ciaran Regan, Sebastian Risi, Jeffrey Seely, Llion Jones

Biological brains demonstrate complex neural activity, where the timing and interplay between neurons is critical to how brains process information. Most deep learning architectures simplify neural activity by abstracting away temporal dynamics. In this paper we challenge that paradigm. By incorporating neuron-level processing and synchronization, we can effectively reintroduce neural timing as a foundational element. We present the Continuous Thought Machine (CTM), a model designed to leverage neural dynamics as its core representation. The CTM has two core innovations: (1) neuron-level temporal processing, where each neuron uses unique weight parameters to process a history of incoming signals; and (2) neural synchronization employed as a latent representation. The CTM aims to strike a balance between oversimplified neuron abstractions that improve computational efficiency, and biological realism. It operates at a level of abstraction that effectively captures essential temporal dynamics while remaining computationally tractable for deep learning. We demonstrate the CTM's strong performance and versatility across a range of challenging tasks, including ImageNet-1K classification, solving 2D mazes, sorting, parity computation, question-answering, and RL tasks. Beyond displaying rich internal representations and offering a natural avenue for interpretation owing to its internal process, the CTM is able to perform tasks that require complex sequential reasoning. The CTM can also leverage adaptive compute, where it can stop earlier for simpler tasks, or keep computing when faced with more challenging instances. The goal of this work is to share the CTM and its associated innovations, rather than pushing for new state-of-the-art results. To that end, we believe the CTM represents a significant step toward developing more biologically plausible and powerful artificial intelligence systems.

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 34 code samples harvested from 2 repositories linked to this paper: 21 ran (of which 9 constructed an object rather than computing a result; 15 with no instrument failure: 0 honoured, 0 violated, 15 with no contract checked; 6 where Syntology's instrument failed) · 13 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 21 samples from 1 repository, 13 ran (11 with no instrument failure; 2 where Syntology's instrument failed); found in paper text by Syntology: 13 samples from 1 repository, 8 ran (4 with no instrument failure; 4 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 21 samples that ran have no recorded date.

SakanaAI/continuous-thought-machines officialmentioned in papermentioned on GitHubpytorchSyntology: 21 samples, 13 ran (2 instrument-failed) · 8 unverified 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.

Code Syntology ran Syntology

34 samples harvested: 21 ran (of which 9 constructed an object rather than computing a result; 15 with no instrument failure: 0 honoured, 0 violated, 15 with no contract checked; 6 where Syntology's instrument failed) · 13 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

5ran · our draft was wrong
1ran; the contract check could not run
15ran
13unverified

Licence: 1 of the 34 samples is pointer only, meaning Syntology does not serve that copy's text. For that sample, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 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. “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 5 samples 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.

SakanaAI/continuous-thought-machines official repository 21 samples, 13 ran (2 instrument-failed) · 8 unverified
CustomRotationalEmbedding models/ctm.py ran fingerprinted Apache-2.0 (permissive) · baaeda06cd538ee1 · report fixture: returned CustomRotationalEmbedding · level recorded 2026-09-22
CustomRotationalEmbedding1D models/ctm.py ran fingerprinted Apache-2.0 (permissive) · 47719b9c89739ef5 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-24
Identity in 49 papers
census: 10 implementations from 13 papers, 4 distinct outputs across 4 buckets, one shared input per bucket
models/ctm.py ran fingerprinted Apache-2.0 (permissive) · e206223b33cedd14 · report fixture: returned Identity · level recorded 2026-09-22
LearnableFourierPositionalEncoding models/ctm.py ran fingerprinted Apache-2.0 (permissive) · e53cf78c38f8573f · report fixture: returned LearnableFourierPositionalEncoding · level recorded 2026-09-22
MultiLearnableFourierPositionalEncoding models/ctm.py ran fingerprinted Apache-2.0 (permissive) · 666bb2b770235116 · report fixture: returned MultiLearnableFourierPositionalEncoding · level recorded 2026-09-22
ParityBackbone models/ctm.py ran fingerprinted Apache-2.0 (permissive) · 90dfef1c76c70aa5 · report fixture: returned ParityBackbone · level recorded 2026-09-22
ResNet in 139 papers models/ctm.py ran Apache-2.0 (permissive) · c67b98fbc8ea482c · report fixture: returned ResNet · level recorded 2026-09-22
ShallowWide models/ctm.py ran Apache-2.0 (permissive) · 5f90461b308ff2c7 · report fixture: returned Tensor · level recorded 2026-09-24
Squeeze in 5 papers models/ctm.py ran Apache-2.0 (permissive) · 36b1c300669d688d · report fixture: returned Squeeze · level recorded 2026-09-22
SuperLinear models/ctm.py ran Apache-2.0 (permissive) · 0461be8a70d9e208 · report fixture: returned SuperLinear · level recorded 2026-09-22
SynapseUNET models/ctm.py ran Apache-2.0 (permissive) · e563747885a2f4ac · report fixture: returned SynapseUNET · level recorded 2026-09-22
add_coord_dim models/ctm.py ran; the contract check could not run (RAISES) Apache-2.0 (permissive) · 73a2570903893c31 · report fixture: signature x:Tensor/3/float, scaled:bool/0/bool · returned Tensor · level recorded 2026-09-22
compute_normalized_entropy models/ctm.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · e9bf59f25bfde7f8 · report fixture: signature logits:Tensor/2/float, reduction:str/0/str · returned Tensor · level recorded 2026-09-22
ContinuousThoughtMachine models/ctm.py unverified Apache-2.0 (permissive) · a6e232630eef5432 · report
_resnet in 27 papers models/ctm.py unverified Apache-2.0 (permissive) · 1c32aa10f477a2b4 · report
prepare_resnet_backbone models/ctm.py unverified Apache-2.0 (permissive) · 213bd7d63b638164 · report
resnet101 in 24 papers models/ctm.py unverified Apache-2.0 (permissive) · 388020d20161fcd4 · report
resnet152 in 5 papers models/ctm.py unverified Apache-2.0 (permissive) · 09646514ded5cb5c · report
resnet18 in 432 papers models/ctm.py unverified Apache-2.0 (permissive) · a1f3dac6cc5b08dd · report
resnet34 in 148 papers models/ctm.py unverified Apache-2.0 (permissive) · 4e5521b72233c2ca · report
resnet50 in 121 papers models/ctm.py unverified Apache-2.0 (permissive) · 9abfa1dee1f9d09c · report
huyvnphan/PyTorch_CIFAR10 found in paper text by Syntology 13 samples, 8 ran (4 instrument-failed) · 5 unverified
conv1x1 in 691 papers cifar10_models/resnet.py ran · our draft was wrong MIT (permissive) · d9def42110729a85 · report fixture: signature in_planes:int/0/int, out_planes:int/0/int, stride:int/0/int · returned Conv2d · level recorded, date not recorded
conv3x3 in 1,376 papers cifar10_models/resnet.py ran · our draft was wrong MIT (permissive) · 160bb14bd76201b4 · report fixture: signature in_planes:int/0/int, out_planes:int/0/int, stride:int/0/int, groups:int/0/int, dilation:int/0/int · returned Conv2d · level recorded, date not recorded
densenet169 in 42 papers cifar10_models/densenet.py ran MIT (permissive) · 9ee8bc3f789e311e · report fixture: signature pretrained:bool/0/bool, progress:bool/0/bool, device:str/0/str · returned DenseNet · level recorded 2026-09-25
make_layers in 155 papers cifar10_models/vgg.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · eba3f5bcc6a06d36 · report fixture: signature cfg:list/1/int, batch_norm:bool/0/bool · returned Sequential · level recorded 2026-09-22
mobilenet_v2 in 23 papers cifar10_models/mobilenetv2.py ran · our draft was wrong MIT (permissive) · 856759d3dba2362b · report fixture: signature pretrained:bool/0/bool, progress:bool/0/bool, device:str/0/str · returned MobileNetV2 · level recorded 2026-09-24
resnet_orig cifar10_models/resnet_orig.py ran MIT (permissive) · cc1974e025ecd3d5 · report fixture: signature pretrained:bool/0/bool, device:str/0/str · returned ResNet · level recorded 2026-09-25
vgg11_bn in 40 papers cifar10_models/vgg.py ran MIT (permissive) · e569c3a2806d4b22 · report fixture: signature pretrained:bool/0/bool, progress:bool/0/bool, device:str/0/str · returned VGG · level recorded 2026-09-25
vgg13_bn cifar10_models/vgg.py ran MIT (permissive) · baf825f46b658a7e · report fixture: signature pretrained:bool/0/bool, progress:bool/0/bool, device:str/0/str · returned VGG · level recorded 2026-09-25
densenet121 in 43 papers cifar10_models/densenet.py unverified MIT (permissive) · d2ef37da818b9f51 · report
densenet161 in 20 papers cifar10_models/densenet.py unverified MIT (permissive) · 2c599b41c3a242e3 · report
googlenet in 8 papers cifar10_models/googlenet.py unverified MIT (permissive) · 2603e42da3f6eb7e · report
inception_v3 in 16 papers cifar10_models/inception.py unverified MIT (permissive) · c042be3fa666e8c8 · report
resnet18 in 432 papers cifar10_models/resnet.py unverified MIT (permissive) · 2615428ce71ead9f · report

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.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for Continuous Thought Machine (CTM) (graph method: Continuous Thought Machine (CTM)). 25 bricks; individually on a fixture: 17 ran (of which 11 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 4 where Syntology's instrument failed) · 8 unverified. The assembly has not been run end-to-end.

  1. ContinuousThoughtMachine entry · level 0 · unverified · a6e232630eef5432
  2. CustomRotationalEmbedding model · level 1 · ran · baaeda06cd538ee1
  3. CustomRotationalEmbedding1D helper · level 1 · ran · 47719b9c89739ef5
  4. LearnableFourierPositionalEncoding model · level 1 · ran · e53cf78c38f8573f
  5. MultiLearnableFourierPositionalEncoding model · level 1 · ran · 666bb2b770235116
  6. ParityBackbone helper · level 1 · ran · 90dfef1c76c70aa5
  7. ShallowWide helper · level 1 · ran · 5f90461b308ff2c7
  8. Squeeze helper · level 1 · ran · 36b1c300669d688d
  9. SuperLinear model · level 1 · ran · 0461be8a70d9e208
  10. SynapseUNET model · level 1 · ran · e563747885a2f4ac
  11. compute_normalized_entropy helper · level 1 · ran · our draft was wrong · e9bf59f25bfde7f8
  12. prepare_resnet_backbone helper · level 0 · unverified · 213bd7d63b638164
  13. add_coord_dim helper · level 1 · ran; the contract check could not run (RAISES) · 73a2570903893c31
  14. Identity helper · level 1 · ran · e206223b33cedd14
  15. resnet101 helper · level 0 · unverified · 388020d20161fcd4
  16. resnet152 helper · level 0 · unverified · 09646514ded5cb5c
  17. resnet18 helper · level 0 · unverified · a1f3dac6cc5b08dd
  18. resnet34 helper · level 0 · unverified · 4e5521b72233c2ca
  19. resnet50 helper · level 0 · unverified · 9abfa1dee1f9d09c
  20. Bottleneck helper · level 1 · ran · 02bba6787a28973f
  21. _resnet helper · level 0 · unverified · 1c32aa10f477a2b4
  22. BasicBlock helper · level 1 · ran · d1351ded2993a927
  23. conv1x1 helper · level 1 · ran · our draft was wrong · d9def42110729a85
  24. conv3x3 helper · level 1 · ran · our draft was wrong · 160bb14bd76201b4
  25. ResNet helper · level 1 · ran · c67b98fbc8ea482c

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="2505.05522") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Computational EfficiencyQuestion Answering

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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