Papers › Autoregressive Video Generation without Vector Quantization

Autoregressive Video Generation without Vector Quantization

18 Dec 2024arXiv:2412.14169archive 2025-07-28

Haoge Deng, Ting Pan, Haiwen Diao, Zhengxiong Luo, Yufeng Cui, Huchuan Lu, Shiguang Shan, Yonggang Qi, Xinlong Wang

This paper presents a novel approach that enables autoregressive video generation with high efficiency. We propose to reformulate the video generation problem as a non-quantized autoregressive modeling of temporal frame-by-frame prediction and spatial set-by-set prediction. Unlike raster-scan prediction in prior autoregressive models or joint distribution modeling of fixed-length tokens in diffusion models, our approach maintains the causal property of GPT-style models for flexible in-context capabilities, while leveraging bidirectional modeling within individual frames for efficiency. With the proposed approach, we train a novel video autoregressive model without vector quantization, termed NOVA. Our results demonstrate that NOVA surpasses prior autoregressive video models in data efficiency, inference speed, visual fidelity, and video fluency, even with a much smaller model capacity, i.e., 0.6B parameters. NOVA also outperforms state-of-the-art image diffusion models in text-to-image generation tasks, with a significantly lower training cost. Additionally, NOVA generalizes well across extended video durations and enables diverse zero-shot applications in one unified model. Code and models are publicly available at https://github.com/baaivision/NOVA.

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 26 code samples harvested from 1 repository linked to this paper: 19 ran (of which 18 constructed an object rather than computing a result; 19 with no instrument failure: 0 honoured, 0 violated, 19 with no contract checked; 0 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 26 samples from 1 repository, 19 ran (19 with no instrument failure; 0 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.

baaivision/nova officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 26 samples, 19 ran (0 instrument-failed) · 7 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

26 samples harvested: 19 ran (of which 18 constructed an object rather than computing a result; 19 with no instrument failure: 0 honoured, 0 violated, 19 with no contract checked; 0 where Syntology's instrument failed) · 7 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.

19ran
7unverified

Licence: 0 of the 26 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 baaivision/nova. “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 11 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.

baaivision/nova official repository 26 samples, 19 ran (0 instrument-failed) · 7 unverified
AdaLayerNorm in 7 papers diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 60c4cd0d5289ebda · report fixture: returned AdaLayerNorm · level recorded 2026-09-22
AdaLayerNormZero diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 85377e46ac705725 · report fixture: returned AdaLayerNormZero · level recorded 2026-09-22
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket
diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · d627a66e082b5f70 · report fixture: returned Attention · level recorded 2026-09-22
Block in 252 papers
census: 5 implementations from 5 papers, 3 distinct outputs across 2 buckets, one shared input per bucket
diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · a6b18aefa9c30394 · report fixture: returned Block · level recorded 2026-09-22
DiffusionBlock diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 9252010cba1de0c1 · report fixture: returned DiffusionBlock · level recorded 2026-09-22
DiffusionMLP diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · c06a16a8febb2446 · report fixture: returned DiffusionMLP · level recorded 2026-09-22
GuidanceScaler diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 028ca8089a24ed40 · report fixture: returned GuidanceScaler · level recorded 2026-09-22
LabelEmbed diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 876ccd77339d9625 · report fixture: returned LabelEmbed · level recorded 2026-09-22
MLP in 375 papers diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 8785a25e456048ac · report fixture: returned MLP · level recorded 2026-09-22
MaskEmbed diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 7a41a8cdb50f38e6 · report fixture: returned MaskEmbed · level recorded 2026-09-22
MotionEmbed diffnext/models/transformers/transformer_nova.py ran fingerprinted Apache-2.0 (permissive) · d1412392f9460cc0 · report fixture: signature c:Tensor/2/float, flow:NoneType/0/?, fps:NoneType/0/? · returned Tensor · level recorded 2026-09-24
PatchEmbed in 123 papers
census: 9 implementations from 9 papers, 6 distinct outputs on one shared input
diffnext/models/transformers/transformer_nova.py ran fingerprinted Apache-2.0 (permissive) · 803f92207c86ec75 · report fixture: returned PatchEmbed · level recorded 2026-09-22
PosEmbed diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 14e66b780fcd44af · report fixture: returned PosEmbed · level recorded 2026-09-22
Projector in 18 papers diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 76ede3fc8a37f868 · report fixture: returned Projector · level recorded 2026-09-22
Registry in 34 papers diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 1f02d40b2da1f5fe · report fixture: returned Registry · level recorded 2026-09-22
RotaryEmbed3D diffnext/models/transformers/transformer_nova.py ran fingerprinted Apache-2.0 (permissive) · 5ec9478ced359867 · report fixture: returned RotaryEmbed3D · level recorded 2026-09-22
TextEmbed diffnext/models/transformers/transformer_nova.py ran fingerprinted Apache-2.0 (permissive) · aa91355939326691 · report fixture: returned TextEmbed · level recorded 2026-09-22
TimeCondEmbed diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · 8103232d2040cad8 · report fixture: returned TimeCondEmbed · level recorded 2026-09-22
VideoPosEmbed diffnext/models/transformers/transformer_nova.py ran Apache-2.0 (permissive) · f36aa3555265a176 · report fixture: returned VideoPosEmbed · level recorded 2026-09-22
FlexAttentionCausal2D diffnext/models/transformers/transformer_nova.py unverified Apache-2.0 (permissive) · ade0129d1a45e0b9 · report
NOVATransformer3DModel diffnext/models/transformers/transformer_nova.py unverified Apache-2.0 (permissive) · 97f6bd68a990166d · report
Transformer3DModel diffnext/models/transformers/transformer_nova.py unverified Apache-2.0 (permissive) · 634c92dbdd87cfe9 · report
VisionTransformer in 53 papers diffnext/models/transformers/transformer_nova.py unverified Apache-2.0 (permissive) · 3a44328c8bf72fe3 · report
image_decoder diffnext/models/transformers/transformer_nova.py unverified Apache-2.0 (permissive) · 7d03ff4b172aacee · report
image_encoder diffnext/models/transformers/transformer_nova.py unverified Apache-2.0 (permissive) · 9dc11e1a48d18afe · report
video_encoder diffnext/models/transformers/transformer_nova.py unverified Apache-2.0 (permissive) · bb1ef5506cb864f5 · 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 NOVA (graph method: NOVA). 26 bricks; individually on a fixture: 19 ran (of which 18 constructed an object rather than computing a result; 19 with no instrument failure: 0 honoured, 0 violated, 19 with no contract checked; 0 where Syntology's instrument failed) · 7 unverified. The assembly has not been run end-to-end.

  1. NOVATransformer3DModel entry · level 0 · unverified · 97f6bd68a990166d
  2. AdaLayerNorm helper · level 1 · ran · 60c4cd0d5289ebda
  3. LabelEmbed helper · level 1 · ran · 876ccd77339d9625
  4. MaskEmbed helper · level 1 · ran · 7a41a8cdb50f38e6
  5. MotionEmbed helper · level 1 · ran · d1412392f9460cc0
  6. PosEmbed helper · level 1 · ran · 14e66b780fcd44af
  7. RotaryEmbed3D helper · level 1 · ran · 5ec9478ced359867
  8. TextEmbed helper · level 1 · ran · aa91355939326691
  9. Transformer3DModel helper · level 0 · unverified · 634c92dbdd87cfe9
  10. VideoPosEmbed helper · level 1 · ran · f36aa3555265a176
  11. image_decoder helper · level 0 · unverified · 7d03ff4b172aacee
  12. image_encoder helper · level 0 · unverified · 9dc11e1a48d18afe
  13. video_encoder helper · level 0 · unverified · bb1ef5506cb864f5
  14. AdaLayerNormZero helper · level 1 · ran · 85377e46ac705725
  15. GuidanceScaler helper · level 1 · ran · 028ca8089a24ed40
  16. DiffusionMLP model · level 1 · ran · c06a16a8febb2446
  17. VisionTransformer model · level 0 · unverified · 3a44328c8bf72fe3
  18. DiffusionBlock helper · level 1 · ran · 9252010cba1de0c1
  19. PatchEmbed helper · level 1 · ran · 803f92207c86ec75
  20. TimeCondEmbed helper · level 1 · ran · 8103232d2040cad8
  21. Block model · level 1 · ran · a6b18aefa9c30394
  22. FlexAttentionCausal2D helper · level 0 · unverified · ade0129d1a45e0b9
  23. Projector helper · level 1 · ran · 76ede3fc8a37f868
  24. Attention model · level 1 · ran · d627a66e082b5f70
  25. MLP model · level 1 · ran · 8785a25e456048ac
  26. Registry helper · level 1 · ran · 1f02d40b2da1f5fe

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="2412.14169") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Image GenerationPredictionQuantizationText to Image GenerationText-to-Image GenerationVideo Generation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Diffusion

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