[whatmodel.sh](../../README.md) / Qwen3-VL 235B-A22B Thinking
# Qwen3-VL 235B-A22B Thinking
Alibaba · released 2025-09 · 235B params (22B active) · apache-2.0
**score 63.3** ±0.7 · #167 of 732 · from 51 benchmark results across 5 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Qwen3-VL 235B-A22B Thinking among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁▃▆█●▅▅▄▄▂▁ #162 of 313 13% 3
coding ▁▁▁▃▅▇█▇●▄▃▁ #137 of 469* 1.6% 1
chat & writing ▁▁▁▃▆▄▆█▆●▇▃ #124 of 402* 6.3% 2
math ▁▁▃▅█▆▄●▄▂▁▁ #131 of 431 13% 6
vision & documents ▁▁▂▂▁▃▄▂●▇█▄ #85 of 153 31% 18
agent tasks · #162 of 313 · 13% of its score
Beats [o1 (2024-12-17)](o1-2024-12-17.md), [o3-mini](o3-mini.md), [DeepSeek R1 0528](deepseek-r1-0528.md) here, though they rank higher overall. Trails [SmolLM3 3B](smollm3-3b.md), [Jamba 1.5 Mini](jamba-1-5-mini.md), [Jamba 1.5 Large](jamba-1-5-large.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
τ²-bench Telecom (maker) 44.7% ▃·█········● beats 4/6 very hard · avg 9.8% 9.3% sharp · at its level
OSWorld-Verified (maker) 38.1% ●▃·▂▅▂▃▂▅▃▆▆ beats 1/27 hard · avg 18% 1.8% at its level · small run
BFCL v3 (maker) 71.8% █▄··▄█·█▄█▄● beats 13/13 medium · avg 59% 1.8% blunt · at its level
coding · #137 of 469 · 1.6% of its score
Beats [Qwen3.5 Plus](qwen3-5-plus.md), [Qwen3-Coder-Next (80B-A3B)](qwen3-coder-next.md), [Gemini 2.5 Pro](gemini-2-5-pro.md) here, though they rank higher overall. Trails [Llama 3.1 Nemotron Ultra 253B v1](llama-3-1-nemotron-ultra-253b.md), [Gemma 4 12B](gemma-4-12b.md), [Seed-OSS 36B Instruct](seed-oss-36b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
LiveCodeBench v6 (Feb–May 2025) (… 70.1% ▁▁▃▁█▅▄●▅▅▃█ beats 21/38 medium · avg 42% 1.6% at its level · small run
chat & writing · #124 of 402 · 6.3% of its score
Beats [LongCat 2.0 (1.6T-A48B)](longcat-2-0.md), [Step 3.5 Flash (196B-A11B)](step-3-5-flash.md), [GPT-5 mini](gpt-5-mini.md) here, though they rank higher overall. Trails [LFM2.5 2.6B](lfm2-5-2-6b.md), [Nemotron-Cascade-8B](nemotron-cascade-8b.md), [DeepSeek V3 0324](deepseek-v3-0324.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
LMArena (text) 1394.7 ▁▁▁▂▃▃▃▆▆●█▃ beats 240/385 avg 1260 5.3% blunt
IFEval (aggregation unspecified) … 88.2% ▁▂▄▂▂▄▆█●▅▂▁ beats 96/145 easy · avg 76% 1.0% small run
math · #131 of 431 · 13% of its score
Beats [Claude Sonnet 4.6](claude-sonnet-4-6.md), [MiniMax-M3](minimax-m3.md), [MiMo-V2.5 (310B-A15B)](mimo-v2-5.md) here, though they rank higher overall. Trails [Olmo 3.1 32B Think](olmo-3-1-32b-think.md), [Qwen3-VL 30B-A3B Thinking](qwen3-vl-30b-a3b-thinking.md), [Granite 4.2 8B](granite-4-2-8b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
AIME 2025 (maker) 89.7% ▁·▁▁▁▂▅▆█▇●▁ beats 115/133 medium · avg 57% 3.5% sharp · small run
HMMT 2025 (edition unspecified) (… 77.4% ▃█▇▅●▅▁·▁··▁ beats 22/32 medium · avg 36% 3.5% sharp · small run
MATH-Vision (maker) 74.6% ●·····▄▄·▄·▄ beats 1/5 medium · avg 47% 2.5% at its level
MathVista (testmini) (maker) 85.8% ▁··▁▄▄▆▇▄▅●▄ beats 41/48 medium · avg 66% 1.6% small run
DynaMath (maker) 82.8% ▂·····▂··●▄▆ beats 3/10 medium · avg 65% 1.2% small run
We-Math (maker) 74.8% ●··········▃ beats 0/3 medium · avg 58% 1.0% small run
vision & documents · #85 of 153 · 31% of its score
Beats [DeepSeek V4 Flash Vision (Exp)](deepseek-v4-flash-vision-exp.md), [GPT-5.4 nano](gpt-5-4-nano.md), [GPT-5 mini](gpt-5-mini.md) here, though they rank higher overall. Trails [GPT-4.1 mini](gpt-4-1-mini.md), [Gemini 2.5 Flash](gemini-2-5-flash.md), [Gemma 4 12B](gemma-4-12b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
CharXiv (reasoning questions) (ma… 66.1% ▄▂▄▆▄▄●··▄█▂ beats 14/21 medium · avg 38% 8.6% at its level
CharXiv descriptive questions (ma… 90.5% ▄··▂▂▂·▂▄▄▂● beats 14/14 easy · avg 74% 4.0% typical
Arena vision (style controlled) 1188.8 ▁▁▂▂▂▂▃▄●▅█▇ beats 67/152 avg 1061 3.5% blunt
MATH-Vision (maker) 74.6% ●·····▄▄·▄·▄ beats 1/5 medium · avg 47% 2.5% at its level
MathVista (testmini) (maker) 85.8% ▁··▁▄▄▆▇▄▅●▄ beats 41/48 medium · avg 66% 1.6% small run
MMMU (val) (maker) 80.6% ▁····▁▁▃█▆▆● beats 47/53 medium · avg 59% 1.5% at its level · small run
MMMU Pro (maker) 69.3% ▂▃▁▃▆▇▃▇●██▅ beats 33/52 medium · avg 47% 1.5% at its level · small run
ChartQA (maker) 90.3% ▂··▂·▂▃·▅▂▆● beats 17/17 easy · avg 85% 1.2% blunt
MMStar (maker) 78.7% ▃▅▂▃▂▂▆██●█▅ beats 27/35 medium · avg 63% 1.2% at its level · small run
ScreenSpot-Pro (maker) 61.8% ▂····▂▄··▂█● beats 10/11 medium · avg 46% 1.2% blunt · at its level
DocVQA (maker) 96.5% ▃··▅▃▅▃█▃█●▅ beats 15/17 very easy · avg 94% 1.0% blunt · maxed out
CountBench (maker) 93.7% ▂▂▃▂·▂▂▃●·▅█ beats 10/18 very easy · avg 89% 1.0% maxed out · small run
SimpleVQA (maker) 61.3% ▃·▃·▃·▃·█●▃▃ beats 8/10 medium · avg 38% 0.9% blunt · at its level · small run
HallusionBench (maker) 66.7% ▂▁·▂▂▂▃▅▃●▅▅ beats 25/33 medium · avg 55% 0.6% blunt · at its level · small run
RealWorldQA (maker) 81.3% ▁▃▁▇▃▇▇▅█●▅▅ beats 31/39 easy · avg 72% 0.4% blunt · small run
CC-OCR (maker) 81.5% ▁·▂··▂▂▃▃▂▆● beats 22/26 easy · avg 72% 0.4% blunt · small run
OCRBench (maker) 87.5% ▁·▃▁▆▅▂█●█▁▃ beats 23/37 easy · avg 83% 0.3% blunt · small run
AI2D test (maker) 89.2% ▁▁▅▇▃▇·●·█·▃ beats 18/28 easy · avg 85% 0.2% blunt · small run
other tests (23)
test result vs the others in its run difficulty counts why
GPQA (subset unspecified) (maker) 77.1% ▂····▂·▂▂█▆● beats 14/14 hard · avg 49% 7.1% sharp · at its level
WebCompass (overall) 38.8% ▅·▃●▃···▃·▃█ beats 3/9 hard · avg 18% 5.3% at its level
AndroidWorld (maker) 62.0% ▃··▃▅▃▃▃▃▃·● beats 9/11 medium · avg 43% 2.9% at its level
Humanity's Last Exam (maker) 13.6% ▁▁▁▄▆●▇██▆▁▃ beats 34/111 very hard · avg 4.4% 2.0% small run
MMLU-Pro (maker) 83.8% ▁▁·▁▁▁▃▂▄▅●▁ beats 160/188 medium · avg 58% 2.0% small run
INCLUDE (maker) 80.0% ▂····▂▂▄▄▄●▄ beats 28/36 medium · avg 58% 1.7% at its level · small run
MMLU-Redux (maker) 93.7% ▁▁▃▁▂▃▄▅▇▅█● beats 58/63 easy · avg 81% 1.7% small run
Creative Writing v3 (maker) 85.7% ▂····▂▆··▆●▆ beats 9/14 easy · avg 67% 1.3% small run
SuperGPQA (maker) 64.3% ▂▁▂▂▅▆▃▇▇▇●▃ beats 39/46 medium · avg 39% 1.2% at its level · small run
VideoMMMU (maker) 80.0% ▂···▂···▂▂●▆ beats 4/10 medium · avg 62% 1.2% small run
MMLU-ProX (maker) 80.6% ▁·▁▁▁▂▁▃▃●▅▄ beats 25/36 medium · avg 59% 1.2% small run
WritingBench (maker) 86.7% ▁·▁▂▁··▅█▂●▁ beats 18/20 easy · avg 71% 1.1% small run
LVBench (maker) 63.6% ▇·█▅●▄▄··▃▁▁ beats 15/28 hard · avg 51% 1.0% at its level · small run
LiveBench 2024-11-25 (maker) 79.6% ▄··▂·▂·▂█▆▂● beats 13/13 medium · avg 59% 1.0% small run
MLVU (maker) 83.8% ▂···▂···▂●▄█ beats 3/10 easy · avg 69% 1.0% small run
MMVU (maker) 71.1% ▃···▃···▃▃●█ beats 4/8 medium · avg 50% 1.0% at its level · small run
RefSpatial-Bench (maker) 69.9% ▃·▃····▃▃▅▅● beats 8/10 medium · avg 45% 0.9% at its level · small run
ERQA (maker) 52.5% ▃··▃··●·▃█▅▅ beats 2/12 medium · avg 36% 0.9% at its level · small run
Multi-IF (aggregation unspecified… 79.1% ▁▁▁▁▁▁▃▄●▄▁▂ beats 29/41 medium · avg 62% 0.9% small run
MMLongBench-Doc (maker) 56.2% ▃···▃···▃●█▅ beats 3/9 medium · avg 37% 0.9% at its level · small run
EmbSpatial-Bench (maker) 84.3% ▁·····▁·▁·▃● beats 7/10 easy · avg 69% 0.8% small run
Video-MME (without subtitles) (ma… 79.0% ▁▃▁▃▁·█▄▄●·▇ beats 21/27 medium · avg 67% 0.8% blunt · at its level · small run
MVBench (maker) 75.2% ▁··▃▄▁▃▄▄▄●▄ beats 22/26 medium · avg 66% 0.4% blunt · at its level · small run
source results share of the score
Qwen (submitted / maker-reported) 9 37% ▇▇▇▇
Qwen/Qwen3.5-397B-A17B model card (subm… 25 25% ▇▇▇
Qwen3-VL-235B-A22B-Thinking model card … 14 24% ▇▇▇
LMArena leaderboard dataset 2 9% ▇
WebCompass: Towards Multimodal Web Codi… 1 5% ▇
prior: pull toward the middle 0% ▇
weighted average = 63.3
Licences: Qwen/Qwen3.5-397B-A17B model card ([apache-2.0](../../licenses/Qwen--Qwen3.5-397B-A17B.txt)) · Qwen3-VL-235B-A22B-Thinking model card ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · WebCompass: Towards Multimodal Web Coding Evaluation for Code Language Models ([CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run Qwen3-VL 235B-A22B Thinking (Q4_K_M, needs ~142.2 GB, from [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF/tree/e3423a389728d101f209c908b79b68367fdfa062))
```sh
llama cli -hf Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF:Q4_K_M
```
Other versions: Q8_0 (~249.9 GB) `llama cli -hf Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF:Q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run Qwen3-VL 235B-A22B Thinking (Q4_K_M, needs ~142.2 GB, from [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF/tree/e3423a389728d101f209c908b79b68367fdfa062))
```sh
lms get https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~249.9 GB) `lms get https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Thinking-GGUF@Q8_0 -y` · 3bit (~104 GB, Mac only) `lms get https://huggingface.co/mlx-community/Qwen3-VL-235B-A22B-Thinking-3bit -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run Qwen3-VL 235B-A22B Thinking (Q4_K_M, needs ~141.5 GB, from [ollama.com](https://ollama.com/library/qwen3-vl/tags))
```sh
ollama run qwen3-vl:235b-a22b-thinking-q4_K_M
```
Other versions: Q8_0 (~282.5 GB) `ollama run qwen3-vl:235b-a22b-thinking-q8_0`
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run Qwen3-VL 235B-A22B Thinking (3bit, needs ~104 GB, Mac only, from [huggingface.co/mlx-community](https://huggingface.co/mlx-community/Qwen3-VL-235B-A22B-Thinking-3bit/tree/ef8a1eef86e928dc5e2de6323ad24ee369050ee5))
```sh
mlx_lm.chat --model mlx-community/Qwen3-VL-235B-A22B-Thinking-3bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model mlx-community/Qwen3-VL-235B-A22B-Thinking-3bit
```
Model repository and documentation: [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Thinking)
[interactive version](https://whatmodel.sh/model/qwen3-vl-235b-a22b-thinking) · [JSON](https://whatmodel.sh/api?format=json&all=1)