Papers › Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models

Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models

7 Nov 2024arXiv:2411.04996archive 2025-07-28

Weixin Liang, Lili Yu, Liang Luo, Srinivasan Iyer, Ning Dong, Chunting Zhou, Gargi Ghosh, Mike Lewis, Wen-tau Yih, Luke Zettlemoyer, Xi Victoria Lin

The development of large language models (LLMs) has expanded to multi-modal systems capable of processing text, images, and speech within a unified framework. Training these models demands significantly larger datasets and computational resources compared to text-only LLMs. To address the scaling challenges, we introduce Mixture-of-Transformers (MoT), a sparse multi-modal transformer architecture that significantly reduces pretraining computational costs. MoT decouples non-embedding parameters of the model by modality -- including feed-forward networks, attention matrices, and layer normalization -- enabling modality-specific processing with global self-attention over the full input sequence. We evaluate MoT across multiple settings and model scales. In the Chameleon 7B setting (autoregressive text-and-image generation), MoT matches the dense baseline's performance using only 55.8% of the FLOPs. When extended to include speech, MoT reaches speech performance comparable to the dense baseline with only 37.2% of the FLOPs. In the Transfusion setting, where text and image are trained with different objectives, a 7B MoT model matches the image modality performance of the dense baseline with one third of the FLOPs, and a 760M MoT model outperforms a 1.4B dense baseline across key image generation metrics. System profiling further highlights MoT's practical benefits, achieving dense baseline image quality in 47.2% of the wall-clock time and text quality in 75.6% of the wall-clock time (measured on AWS p4de.24xlarge instances with NVIDIA A100 GPUs).

PaperPDFCodeCode Syntology ran

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 harvested from 1 repository linked to this paper: 12 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 1 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: community (archive-listed): 13 samples from 1 repository, 12 ran (10 with no instrument failure; 2 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 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

allenzren/open-pi-zero mentioned on GitHubpytorchMITSyntology: 13 samples, 12 ran (2 instrument-failed) · 1 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

13 samples harvested: 12 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 2 where Syntology's instrument failed) · 1 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.

1ran · honoured contract
1ran · our draft was wrong
1ran; the contract check could not run
9ran
1unverified

Licence: 2 of the 13 samples are pointer only, meaning Syntology does not serve that copy's text. For 2 of them, 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 allenzren/open-pi-zero. “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”. “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.

allenzren/open-pi-zero community (archive-listed) 13 samples, 12 ran (2 instrument-failed) · 1 unverified
conditional_decorator src/utils/decorator.py ran · our draft was wrong MIT (permissive) · d949d0d0bc5e63bc · report fixture: signature dec:callable/0/?, condition:bool/0/bool · returned function · level recorded 2026-09-24
get_action_accuracy src/utils/metric.py ran MIT (permissive) · 7f39687304604d5d · report fixture: signature gt:Tensor/2/float, pred:Tensor/2/float, thresholds:list/1/float · returned Tensor · level recorded 2026-09-24
get_num_params_in_billions src/utils/optim.py ran MIT (permissive) · c309b4c8e3254bb9 · report fixture: signature optimizer:SGD/0/? · returned float · level recorded 2026-09-24
isrotation src/utils/geometry.py ran MIT (permissive) · ee388693e9bdd485 · report fixture: signature R:ndarray/2/float, thresh:float/0/float · returned bool · level recorded 2026-09-24
log_execution_time src/utils/monitor.py ran MIT (permissive) · 1c57d6dfeacfc394 · report fixture: signature logger:Logger/0/? · returned function · level recorded 2026-09-24
lora_state_dict in 23 papers src/model/lora.py ran MIT (permissive) · 52614cdfc99cf150 · report fixture: signature model:callable/0/?, train_bias:str/0/str, train_head:bool/0/bool · returned dict · level recorded 2026-09-24
main_rank_only src/utils/decorator.py ran MIT (permissive) · 8d01c3eb9b8242f3 · report fixture: signature func:callable/0/? · returned function · level recorded 2026-09-24
mat2euler in 10 papers src/utils/geometry.py ran MIT (permissive) · eed6f0c4289630b1 · report fixture: signature mat:ndarray/2/float, axes:str/0/str · returned tuple · level recorded 2026-09-24
quat2mat in 26 papers
census: 5 implementations from 5 papers, 3 distinct outputs across 2 buckets, one shared input per bucket
src/utils/geometry.py ran MIT (permissive) · 901875b18f961eef · report fixture: signature q:list/1/int · returned ndarray · level recorded 2026-09-24
repeat_kv in 244 papers
census: 15 implementations from 60 papers, 2 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 4, kind float, dtype float32
src/model/utils.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) here; the stored copy is pointer only · 4e762613ee907cdb · report fixture: signature hidden_states:Tensor/4/float, n_rep:int/0/int · returned Tensor · level recorded 2026-09-22
rotate_half in 424 papers
census: 24 implementations from 96 papers, 4 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 2, kind float, dtype float32
src/model/utils.py ran · honoured contract fingerprinted MIT (permissive) here; the stored copy is pointer only · cec833a99feb528c · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
transpose in 53 papers
census: 6 implementations from 13 papers, 3 distinct outputs across 2 buckets, one shared input per bucket
src/model/lora.py ran fingerprinted MIT (permissive) · 67a5b7072d402f6f · report fixture: signature weight:Tensor/2/float, fan_in_fan_out:bool/0/bool · returned Tensor · level recorded 2026-09-24

Tasks

Image Generation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AttentionLayer NormalizationSoftmax

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