Papers › Emu3: Next-Token Prediction is All You Need

Emu3: Next-Token Prediction is All You Need

27 Sep 2024arXiv:2409.18869archive 2025-07-28

Xinlong Wang, Xiaosong Zhang, Zhengxiong Luo, Quan Sun, Yufeng Cui, Jinsheng Wang, Fan Zhang, Yueze Wang, Zhen Li, Qiying Yu, Yingli Zhao, Yulong Ao, Xuebin Min, Tao Li, Boya Wu, Bo Zhao, BoWen Zhang, Liangdong Wang, Guang Liu, Zheqi He, Xi Yang, Jingjing Liu, Yonghua Lin, Tiejun Huang, Zhongyuan Wang

While next-token prediction is considered a promising path towards artificial general intelligence, it has struggled to excel in multimodal tasks, which are still dominated by diffusion models (e.g., Stable Diffusion) and compositional approaches (e.g., CLIP combined with LLMs). In this paper, we introduce Emu3, a new suite of state-of-the-art multimodal models trained solely with next-token prediction. By tokenizing images, text, and videos into a discrete space, we train a single transformer from scratch on a mixture of multimodal sequences. Emu3 outperforms several well-established task-specific models in both generation and perception tasks, surpassing flagship models such as SDXL and LLaVA-1.6, while eliminating the need for diffusion or compositional architectures. Emu3 is also capable of generating high-fidelity video via predicting the next token in a video sequence. We simplify complex multimodal model designs by converging on a singular focus: tokens, unlocking great potential for scaling both during training and inference. Our results demonstrate that next-token prediction is a promising path towards building general multimodal intelligence beyond language. We open-source key techniques and models to support further research in this direction.

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2409.18869")

Code

Syntology Ran 3 of 6 code samples harvested from 1 repository linked to this paper; 3 have no recorded run. Of those that ran: 1 ran · our draft was wrong; 2 ran · fixture could not drive it.

By repository: community (archive-listed): 6 samples from 1 repository, 3 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

baaivision/emu3 mentioned on GitHubpytorchApache-2.0 report
flagopen/flagscale mentioned on GitHubpytorchApache-2.0 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

6 samples harvested; 3 ran; 0 honoured the contract we drafted; 3 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · our draft was wrong
2ran · fixture could not drive it
3unverified

Licence: 0 of the 6 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/emu3. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

repeat_kv baaivision/emu3/emu3/mllm/modeling_emu3.py community (archive-listed) ran · fixture could not drive it fingerprinted Apache-2.0 (permissive) · 30d7eec482ebf6b1 · report
apply_rotary_pos_emb baaivision/emu3/emu3/mllm/modeling_emu3.py community (archive-listed) ran · fixture could not drive it Apache-2.0 (permissive) · d61c483a3c2b3156 · report
rotate_half baaivision/emu3/emu3/mllm/modeling_emu3.py community (archive-listed) ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · b99eea6376d1e212 · report
image2str baaivision/emu3/gradio_demo.py community (archive-listed) unverified Apache-2.0 (permissive) · 658cb59bfa54f294 · report
smart_resize baaivision/emu3/emu3/tokenizer/image_processing_emu3visionvq.py community (archive-listed) unverified Apache-2.0 (permissive) · 5ea70bedcc69c330 · report
smart_resize baaivision/emu3/emu3/train/prepare_data.py community (archive-listed) unverified Apache-2.0 (permissive) · 8d1dbae469f6cafb · report

Tasks

AllImage GenerationPredictionText-to-Image GenerationVisual Question Answering

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
WISE Emu3 Biology 0.41 #8 of 11 Archive leaderboard report
WISE Emu3 Chemistry 0.27 #8 of 11 Archive leaderboard report
WISE Emu3 Cultural 0.34 #8 of 11 Archive leaderboard report
WISE Emu3 Overall 0.39 #8 of 11 Archive leaderboard report
WISE Emu3 Physics 0.45 #8 of 11 Archive leaderboard report
WISE Emu3 Space 0.48 #8 of 11 Archive leaderboard report
WISE Emu3 Time 0.45 #8 of 11 Archive leaderboard report
Image Generation WISE Emu3-gen Biology 0.41 #11 of 14 Archive leaderboard report
Image Generation WISE Emu3-gen Chemistry 0.27 #11 of 14 Archive leaderboard report
Image Generation WISE Emu3-gen Cultural 0.34 #11 of 14 Archive leaderboard report
Image Generation WISE Emu3-gen Overall 0.39 #11 of 14 Archive leaderboard report
Image Generation WISE Emu3-gen Physics 0.45 #11 of 14 Archive leaderboard report
Image Generation WISE Emu3-gen Space 0.48 #11 of 14 Archive leaderboard report
Image Generation WISE Emu3-gen Time 0.45 #11 of 14 Archive leaderboard report
Text-to-Image Generation GenEval Emu3 Overall 0.66 #15 of 20 Archive leaderboard report
Text-to-Image Generation T2I-CompBench Emu3 Color 0.7913 #1 of 2 Archive leaderboard report
Text-to-Image Generation T2I-CompBench Emu3 Shape 0.5846 #1 of 2 Archive leaderboard report
Text-to-Image Generation T2I-CompBench Emu3 Texture 0.7422 #1 of 2 Archive leaderboard report
Visual Question Answering MM-Vet Emu3 GPT-4 score 37.2 #138 of 231 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

CLIPDiffusion

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