Перейти к основному содержанию

Qwen3 VL 235B A22B (Reasoning)

AlibabaQwenОткрытые весаApache 2.0 · Коммерческое использование

Описание

Qwen3-VL-235B-A22B-Thinking is the most powerful vision-language model in the Qwen series, featuring 236B parameters with MoE architecture for reasoning-enhanced multimodal understanding. Key capabilities include: Visual Agent (operates PC/mobile GUIs, recognizes elements, invokes tools), Visual Coding (generates Draw.io/HTML/CSS/JS from images/videos), Advanced Spatial Perception (2D grounding and 3D grounding for spatial reasoning and embodied AI), Long Context & Video Understanding (native 256K context expandable to 1M, handles hours-long video with second-level indexing), Enhanced Multimodal Reasoning (excels in STEM/Math with causal analysis), Upgraded Visual Recognition (celebrities, anime, products, landmarks, flora/fauna), and Expanded OCR (32 languages, robust in low light/blur/tilt). Architecture innovations include Interleaved-MRoPE for positional embeddings, DeepStack for multi-level ViT feature fusion, and Text-Timestamp Alignment for precise video temporal modeling.

Дата выхода
2025-09-23
Параметры
236.0B
Длина контекста
131K
Модальности
image, text, video

Радар способностей

34
general
65
coding
86
reasoning
57
science
70
agents
100
multimodal

Рейтинги

Домен#МестоОценкаИсточник
Рейтинг кодинга274
53.0
AA
Общий рейтинг231
50.0
AA
Мультимодальный рейтинг81
51.0
LS
Наука279
48.0
AA

Оценки бенчмарков (LLM Stats)

(LLM Stats (zeroeval))

Chat

MM-MT-Bench8.50 / 100Сам.
IFEvalGoogle Research (2023)88.2%Сам.
Multi-IF79.1%Сам.

Creativity

Creative Writing v385.7%Сам.

Factuality

SimpleQA44.4%Сам.

General

MMLU90.6%Сам.
Include80.0%Сам.
SIFO0.77 / 100Сам.
BFCL-v371.9%Сам.
SIFO-Multiturn0.71 / 100Сам.

Language

MMLU-Redux93.7%Сам.
MMLU-Pro83.8%Сам.
MMLU-ProX80.6%Сам.

Math

AIME 202589.7%Сам.
MathVista-Mini85.8%Сам.
MathVerse-Mini0.85 / 100Сам.
LiveBench 2024112579.6%Сам.
HMMT2577.4%Сам.
MathVision74.6%Сам.

Multimodal

VideoMMMU80.0%Сам.
VideoMME w/o sub.79.0%Сам.
OSWorld-G0.68 / 100Сам.
OSWorld38.1%Сам.

Reasoning

ZebraLogic97.3%Сам.
LiveCodeBench v670.1%Сам.
CharXiv-R66.1%Сам.
SuperGPQA64.3%Сам.
Humanity's Last Exam13.6%Сам.

Video

MLVU83.8%Сам.
CharadesSTA63.5%Сам.

Vision

DocVQAtest96.5%Сам.
ScreenSpot95.4%Сам.
CountBench0.94 / 100Сам.
Design2Code0.93 / 100Сам.
MIABench0.93 / 100Сам.
RefCOCO-avg0.92 / 100Сам.
MMBench-V1.190.6%Сам.
InfoVQAtest89.5%Сам.
AI2D89.2%Сам.
OCRBench87.5%Сам.
EmbSpatialBench0.84 / 100Сам.
CC-OCR81.5%Сам.
RealWorldQA81.3%Сам.
MMMUval80.6%Сам.
MuirBench80.1%Сам.
MMStar78.7%Сам.
RoboSpatialHome0.74 / 100Сам.
Objectron0.71 / 100Сам.
RefSpatialBench0.70 / 100Сам.
MMMU-Pro69.3%Сам.
BLINK67.1%Сам.
OCRBench-V2 (en)66.8%Сам.
Hallusion Bench66.7%Сам.
LVBench63.6%Сам.
OCRBench-V2 (zh)63.5%Сам.
ScreenSpot Pro61.8%Сам.
SimpleVQA0.61 / 100Сам.
MMLongBench-Doc0.56 / 100Сам.
ARKitScenes0.54 / 100Сам.
ERQA52.5%Сам.
ODinW43.2%Сам.
SUNRGBD0.35 / 100Сам.
VisuLogic0.34 / 100Сам.
ZEROBench-Sub0.28 / 100Сам.
Hypersim0.11 / 100Сам.
ZEROBench0.04 / 100Сам.

Writing

WritingBench86.7%Сам.

Индексы оценки AA

(Artificial Analysis)
Aime 25(MAA (Mathematical Association of America))
88.3
Math Index(Artificial Analysis)
88.3
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
83.6
Gpqa(NYU + Cohere + Anthropic (2023))
77.2
Livecodebench(UC Berkeley + MIT + Cornell (2024))
64.6
Lcr(Artificial Analysis)
63.7
Ifbench(Google Research (2023))
56.5
Tau2(Sierra + U Toronto + Vector Institute (2025))
54.1
Intelligence Index(Artificial Analysis)
13.4
Hle(Center for AI Safety + Scale AI (2025))
11.9
Terminalbench Hard(Stanford × Laude Institute (2026))
11.4

Оценки категорий LLM Stats

(LLM Stats (zeroeval))
Chat
3
Communication
3
Multimodal
100
Creativity
90
Writing
90
Instruction Following
80
Language
80
Legal
80
Math
80
Structured Output
80
Embodied
80
Finance
80
General
80
Grounding
80
Healthcare
80
Text-to-image
80
Video
80
Image To Text
70
Long Context
70
Reasoning
70
Spatial Reasoning
70
Tool Calling
70
Vision
70
Physics
60
Agents
60
Chemistry
60
Economics
60
Factuality
40
3d
40

Цены

Цена ввода$0.4 / 1M токенов
Цена вывода$4 / 1M токенов
Смешанная цена (3:1)$1.3 / 1M токенов

Скорость

Токенов/сек0.0
Задержка первого токена0.00s
Время до первого ответа0.00s

Рейтинг цен провайдеров

Рейтинг цен провайдеров

11 провайдеров

Самый дешевый: Venice AIСамый дорогой: NovitaAI
ПровайдерВводВывод
1Venice AIСамый дешевый
$0.21
$1.9
2Alibaba (China)
$0.28671
$1.14682
3Merge Gateway
$0.287
$2.867
4AlibabaОсновной
$0.4
$4
5OpenRouter
$0.4
$4
6Kilo Gateway
$0.4
$4
7OrcaRouter
$0.4
$4
8Eden AI
$0.4
$4
9NanoGPT
$0.5
$6
10Cortecs
$0.617
$3.119
11NovitaAI
$0.98
$3.95

Сравнение цен разных API-провайдеров для этой модели.

Внешние ссылки