Papers › Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution

Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution

18 Sep 2024arXiv:2409.12191archive 2025-07-28

Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Yang Fan, Kai Dang, Mengfei Du, Xuancheng Ren, Rui Men, Dayiheng Liu, Chang Zhou, Jingren Zhou, Junyang Lin

We present the Qwen2-VL Series, an advanced upgrade of the previous Qwen-VL models that redefines the conventional predetermined-resolution approach in visual processing. Qwen2-VL introduces the Naive Dynamic Resolution mechanism, which enables the model to dynamically process images of varying resolutions into different numbers of visual tokens. This approach allows the model to generate more efficient and accurate visual representations, closely aligning with human perceptual processes. The model also integrates Multimodal Rotary Position Embedding (M-RoPE), facilitating the effective fusion of positional information across text, images, and videos. We employ a unified paradigm for processing both images and videos, enhancing the model's visual perception capabilities. To explore the potential of large multimodal models, Qwen2-VL investigates the scaling laws for large vision-language models (LVLMs). By scaling both the model size-with versions at 2B, 8B, and 72B parameters-and the amount of training data, the Qwen2-VL Series achieves highly competitive performance. Notably, the Qwen2-VL-72B model achieves results comparable to leading models such as GPT-4o and Claude3.5-Sonnet across various multimodal benchmarks, outperforming other generalist models. Code is available at https://github.com/QwenLM/Qwen2-VL .

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 12 code samples harvested from 4 repositories linked to this paper: 12 ran (of which 1 constructed an object rather than computing a result; 10 with no instrument failure: 4 honoured, 1 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 4 samples from 1 repository, 4 ran (4 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 8 samples from 3 repositories, 8 ran (6 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.

No sample is linked to this paper from 2 of its repositories that Syntology harvested for other papers:

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 12 samples that ran have no recorded date.

qwenlm/qwen2-vl officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 4 samples, 4 ran (0 instrument-failed) · 0 unverified report
baichuan-inc/Baichuan-Omni-1.5 mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
juruobenruo/DexVLA mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
qwenlm/qwen2.5-vl mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ACL Anthology id 2025.findings-acl.964 (that paper's own run record: 17 ran (of which 0 constructed an object rather than computing a result; 14 with no instrument failure: 1 honoured, 0 violated, 13 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
tutujingyugang1/ChatVLA_public mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ChatVLA-2: Vision-Language-Action Model with Open-World Embodied Reasoning from... (arXiv:2505.21906; that paper's own run record: 13 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified); nothing checks that they implement this paper's method. report
MindCode-4/code-4 mindsporeSyntology: 2 samples, 2 ran (2 instrument-failed) · 0 unverified report
MindCode-4/code-4 mindsporeSyntology: 2 samples, 2 ran (2 instrument-failed) · 0 unverified report
yangyucheng000/University mindsporeSyntology: 5 samples, 5 ran (0 instrument-failed) · 0 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

12 samples harvested: 12 ran (of which 1 constructed an object rather than computing a result; 10 with no instrument failure: 4 honoured, 1 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 0 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.

4ran · honoured contract
1ran · violated contract
2ran · our draft was wrong
5ran

Licence: 1 of the 12 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 4 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”. “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.

qwenlm/qwen2-vl official repository 4 samples, 4 ran (0 instrument-failed) · 0 unverified
ceil_by_factor in 24 papers qwen-vl-utils/src/qwen_vl_utils/vision_process.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 6e45201fa27cb24a · report fixture: signature number:int/0/int, factor:int/0/int · returned int · level recorded 2026-09-22
floor_by_factor in 24 papers qwen-vl-utils/src/qwen_vl_utils/vision_process.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 8155263d7ff19bb3 · report fixture: signature number:int/0/int, factor:int/0/int · returned int · level recorded 2026-09-22
round_by_factor in 26 papers qwen-vl-utils/src/qwen_vl_utils/vision_process.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · e252767324188623 · report fixture: signature number:int/0/int, factor:int/0/int · returned int · level recorded 2026-09-22
smart_resize in 11 papers qwen-vl-utils/src/qwen_vl_utils/vision_process.py ran · honoured contract Apache-2.0 (permissive) · 464fd25cf4b2eac6 · report fixture: signature height:int/0/int, width:int/0/int, factor:int/0/int, min_pixels:NoneType/0/?, max_pixels:NoneType/0/? · returned tuple · level recorded 2026-09-22
baichuan-inc/Baichuan-Omni-1.5 community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
RotaryEmbedding in 27 papers baichuan-omni/model/modeling_omni.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 7bc2a2784aabedc7 · report fixture: returned RotaryEmbedding · level recorded 2026-09-22
MindCode-4/code-4 community (archive-listed) 2 samples, 2 ran (2 instrument-failed) · 0 unverified
whitespace_tokenize in 49 papers prophetnet/tokenization_prophetnet.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · cf9ffa02a42184af · report fixture: signature text:str/0/str · returned list · level recorded, date not recorded
load_vocab in 107 papers prophetnet/tokenization_prophetnet.py ran · our draft was wrong Apache-2.0 (permissive) · e7fbc7a74a3457c7 · report fixture: signature vocab_file:str/0/str · returned OrderedDict · level recorded, date not recorded
yangyucheng000/University community (archive-listed) 5 samples, 5 ran (0 instrument-failed) · 0 unverified
flow_color JDRL-mindspore/flowfunction.py ran Apache-2.0 (permissive) · 6ccf5863fa0f0c9d · report fixture: signature flow:MockFlowTensor/0/? · returned ndarray · level recorded 2026-09-24
flow_compute_color in 7 papers JDRL-mindspore/flowfunction.py ran Apache-2.0 (permissive) · 63c5a2b51dee1425 · report fixture: signature u:ndarray/2/float, v:ndarray/2/float, convert_to_bgr:bool/0/bool · returned ndarray · level recorded 2026-09-24
flow_to_color in 10 papers JDRL-mindspore/flowfunction.py ran Apache-2.0 (permissive) · c869c75bbde6fc92 · report fixture: signature flow_uv:ndarray/3/float, clip_flow:float/0/float, convert_to_bgr:bool/0/bool · returned ndarray · level recorded 2026-09-25
is_image_file in 157 papers JDRL-mindspore/dataset_RGB.py ran · violated contract Apache-2.0 (permissive) here; the stored copy is pointer only · 91fca7a2b44569ad · report fixture: signature filename:str/0/str · returned bool · level recorded 2026-09-24
train in 522 papers PDF_MS/trainner.py ran Apache-2.0 (permissive) · 2fe12723d2eed03d · report fixture: signature model:callable/0/?, dataloader:list/1/Batch, train_net:callable/0/? · returned list · level recorded 2026-09-24

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.

2 assemblies on this paper: 2 harvested, 0 generated.

harvested Assembly for Multimodal Rotary Position Embedding (M-RoPE) (graph method: Multimodal Rotary Position Embedding (M-RoPE)). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

  1. RotaryEmbedding entry · level 1 · ran · 7bc2a2784aabedc7

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

  1. smart_resize entry · level 1 · ran · honoured contract · 464fd25cf4b2eac6
  2. ceil_by_factor helper · level 1 · ran · honoured contract · 6e45201fa27cb24a
  3. floor_by_factor helper · level 1 · ran · honoured contract · 8155263d7ff19bb3
  4. round_by_factor helper · level 1 · ran · honoured contract · e252767324188623

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="2409.12191") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Natural Language Visual GroundingTemporal Relation ExtractionVideo Question AnsweringVisual Question AnsweringVisual Question Answering (VQA)Zero-Shot Video Question Answer

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
ImplicitQA Qwen2 VL - 7B Average Accuracy 44.9 #3 of 7 Archive leaderboard report
ImplicitQA Qwen2 VL - 7B Macro Average Accuracy 46.0 #3 of 7 Archive leaderboard report
Natural Language Visual Grounding ScreenSpot Qwen2-VL-7B Accuracy (%) 42.1 #14 of 18 Archive leaderboard report
Temporal Relation Extraction Vinoground Qwen2-VL-72B Group Score 17.4 #3 of 24 Archive leaderboard report
Temporal Relation Extraction Vinoground Qwen2-VL-72B Text Score 50.4 #3 of 24 Archive leaderboard report
Temporal Relation Extraction Vinoground Qwen2-VL-72B Video Score 32.6 #3 of 24 Archive leaderboard report
Temporal Relation Extraction Vinoground Qwen2-VL-7B Group Score 15.2 #6 of 24 Archive leaderboard report
Temporal Relation Extraction Vinoground Qwen2-VL-7B Text Score 40.2 #6 of 24 Archive leaderboard report
Temporal Relation Extraction Vinoground Qwen2-VL-7B Video Score 32.4 #6 of 24 Archive leaderboard report
Video Question Answering NExT-QA Qwen2-VL(7B) Accuracy 81.2 #10 of 47 Archive leaderboard report
Video Question Answering OVBench Qwen2-VL (7B) AVG 49.7 #4 of 16 Archive leaderboard report
Video Question Answering TVBench Qwen2-VL-72B Average Accuracy 52.7 #9 of 28 Archive leaderboard report
Video Question Answering TVBench Qwen2-VL-7B Average Accuracy 43.8 #18 of 28 Archive leaderboard report
Visual Question Answering MM-Vet Qwen2-VL-72B GPT-4 score 74.0 #6 of 231 Archive leaderboard report
Visual Question Answering MM-Vet Qwen2-VL-7B GPT-4 score 62.0 #32 of 231 Archive leaderboard report
Visual Question Answering MM-Vet Qwen2-VL-2B GPT-4 score 49.5 #68 of 231 Archive leaderboard report
Visual Question Answering MM-Vet v2 Qwen2-VL-72B (qwen-vl-max-0809) GPT-4 score 66.9±0.3 #7 of 24 Archive leaderboard report
Visual Question Answering MM-Vet v2 Qwen2-VL-72B (qwen-vl-max-0809) Params 72B #7 of 24 Archive leaderboard report
Visual Question Answering (VQA) VLM2-Bench Qwen2-VL-7B Average Score on VLM2-bench (9 subtasks) 42.37 #4 of 9 Archive leaderboard report
Visual Question Answering (VQA) VLM2-Bench Qwen2-VL-7B GC-mat 27.80 #4 of 9 Archive leaderboard report
Visual Question Answering (VQA) VLM2-Bench Qwen2-VL-7B GC-trk 19.18 #4 of 9 Archive leaderboard report
Visual Question Answering (VQA) VLM2-Bench Qwen2-VL-7B OC-cnt 45.99 #4 of 9 Archive leaderboard report
Visual Question Answering (VQA) VLM2-Bench Qwen2-VL-7B OC-cpr 68.06 #4 of 9 Archive leaderboard report
Visual Question Answering (VQA) VLM2-Bench Qwen2-VL-7B OC-grp 35.00 #4 of 9 Archive leaderboard report
Visual Question Answering (VQA) VLM2-Bench Qwen2-VL-7B PC-VID 16.25 #4 of 9 Archive leaderboard report
Visual Question Answering (VQA) VLM2-Bench Qwen2-VL-7B PC-cnt 58.59 #4 of 9 Archive leaderboard report
Visual Question Answering (VQA) VLM2-Bench Qwen2-VL-7B PC-cpr 61.50 #4 of 9 Archive leaderboard report
Visual Question Answering (VQA) VLM2-Bench Qwen2-VL-7B PC-grp 49.00 #4 of 9 Archive leaderboard report
Zero-Shot Video Question Answer VNBench Qwen2-VL-7B Accuracy 33.9 #5 of 9 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.

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