Papers › DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding

DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding

13 Dec 2024arXiv:2412.10302archive 2025-07-28

Zhiyu Wu, Xiaokang Chen, Zizheng Pan, Xingchao Liu, Wen Liu, Damai Dai, Huazuo Gao, Yiyang Ma, Chengyue Wu, Bingxuan Wang, Zhenda Xie, Yu Wu, Kai Hu, Jiawei Wang, Yaofeng Sun, Yukun Li, Yishi Piao, Kang Guan, Aixin Liu, Xin Xie, Yuxiang You, Kai Dong, Xingkai Yu, Haowei Zhang, Liang Zhao, Yisong Wang, Chong Ruan

We present DeepSeek-VL2, an advanced series of large Mixture-of-Experts (MoE) Vision-Language Models that significantly improves upon its predecessor, DeepSeek-VL, through two key major upgrades. For the vision component, we incorporate a dynamic tiling vision encoding strategy designed for processing high-resolution images with different aspect ratios. For the language component, we leverage DeepSeekMoE models with the Multi-head Latent Attention mechanism, which compresses Key-Value cache into latent vectors, to enable efficient inference and high throughput. Trained on an improved vision-language dataset, DeepSeek-VL2 demonstrates superior capabilities across various tasks, including but not limited to visual question answering, optical character recognition, document/table/chart understanding, and visual grounding. Our model series is composed of three variants: DeepSeek-VL2-Tiny, DeepSeek-VL2-Small and DeepSeek-VL2, with 1.0B, 2.8B and 4.5B activated parameters respectively. DeepSeek-VL2 achieves competitive or state-of-the-art performance with similar or fewer activated parameters compared to existing open-source dense and MoE-based models. Codes and pre-trained models are publicly accessible at https://github.com/deepseek-ai/DeepSeek-VL2.

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="2412.10302")

Code

Syntology Ran 4 of 13 code samples harvested from 1 repository linked to this paper; 9 have no recorded run. Of those that ran: 1 ran · our draft was wrong; 3 ran with no contract checked.

By repository: official repository: 13 samples from 1 repository, 4 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

deepseek-ai/deepseek-vl2 officialmentioned in papermentioned on GitHubpytorchMIT 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; 4 ran; 0 honoured the contract we drafted; 9 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
3ran
9unverified

Licence: 0 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 deepseek-ai/deepseek-vl2. “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.

load_json deepseek-ai/deepseek-vl2/deepseek_vl2/utils/io.py official repository ran · our draft was wrong MIT (permissive) · 96f3e857e91a8cb8 · report
yarn_find_correction_dim deepseek-ai/deepseek-vl2/deepseek_vl2/models/modeling_deepseek.py official repository ran fingerprinted MIT (permissive) · de3c1ddb7a5bbb47 · report
yarn_find_correction_range deepseek-ai/deepseek-vl2/deepseek_vl2/models/modeling_deepseek.py official repository ran MIT (permissive) · ed5f1f4e8c0439b2 · report
yarn_get_mscale deepseek-ai/deepseek-vl2/deepseek_vl2/models/modeling_deepseek.py official repository ran fingerprinted MIT (permissive) · 8dfdb3a53834d90f · report
compact_text_chunks deepseek-ai/deepseek-vl2/deepseek_vl2/serve/app_modules/overwrites.py official repository unverified MIT (permissive) · ba86a364bce26e9b · report
create_siglip_vit deepseek-ai/deepseek-vl2/deepseek_vl2/models/siglip_vit.py official repository unverified MIT (permissive) · b675d2f1fa15bd63 · report
format_output deepseek-ai/deepseek-vl2/deepseek_vl2/serve/app_modules/utils.py official repository unverified MIT (permissive) · 75628f3739d6fcd3 · report
get_conv_template deepseek-ai/deepseek-vl2/deepseek_vl2/models/conversation.py official repository unverified MIT (permissive) · 132894d1028abd5a · report
load_pil_images deepseek-ai/deepseek-vl2/deepseek_vl2/utils/io.py official repository unverified MIT (permissive) · 56ce205e9245c8c2 · report
load_pretrained_model deepseek-ai/deepseek-vl2/deepseek_vl2/utils/io.py official repository unverified MIT (permissive) · 8a1727ee6e04ec45 · report
markdown_to_html_with_syntax_highlight deepseek-ai/deepseek-vl2/deepseek_vl2/serve/app_modules/utils.py official repository unverified MIT (permissive) · d1fcc287229a99c6 · report
select_best_resolution deepseek-ai/deepseek-vl2/deepseek_vl2/models/processing_deepseek_vl_v2.py official repository unverified MIT (permissive) · 6408ebfc6065bf19 · report
strip_stop_words deepseek-ai/deepseek-vl2/deepseek_vl2/serve/app_modules/utils.py official repository unverified MIT (permissive) · f7fc8e891d83a3e1 · report

Tasks

Chart UnderstandingMixture-of-ExpertsOptical Character RecognitionQuestion AnsweringReferring Expression ComprehensionVisual GroundingVisual Question Answering

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AttentionSoftmax

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