whatmodel.sh / Qwen3-VL 30B-A3B Instruct

Qwen3-VL 30B-A3B Instruct

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Qwen3-VL 30B-A3B Instruct is an open-weights language model from Alibaba, released September 2025. Its published size is 31.1B parameters (3B active per token) and a 256k-token context window. Licence: apache-2.0.

Catalogue identity notes

Maker repo Qwen/Qwen3-VL-30B-A3B-Instruct @9c4b90e (licence tag apache-2.0, created 2025-09-30); the separate Instruct checkpoint, not Qwen3-VL-30B-A3B-Thinking. Scored from the Qwen3-VL Technical Report (arXiv 2511.21631v2, Tables 3/4 and 7-10: column 'Qwen3-VL 30B-A3B instruct'), whose instruct and thinking columns are the two published checkpoints. Released 2025-09: month the maker's Hugging Face repo was created.

Model information sources

Release date ↗ · Weights licence ↗

31,070,754,032 parameters in the Hugging Face safetensors metadata of Qwen/Qwen3-VL-30B-A3B-Instruct @9c4b90e (vision encoder included); MoE with ~3B activated per token per the model name ('30B-A3B').

57.8±0.8#221 of 719 LLMsfrom 25 benchmark results across 1 source

$ how it compares

The LLMs ranked either side of Qwen3-VL 30B-A3B Instruct, out of 828 with a score, then more from Alibaba.

#modelscorewhereRAMreleased
263Magistral Small 1.2 Mistral58±1api $0.75/M +local~16GB2025-09
264Nemotron-Cascade-8B-Thinking NVIDIA58±2local~6.3GB2025-12
265Qwen3-VL 30B-A3B Instruct Alibaba58±1api $0.35/M +local~21GB2025-09
266Qwen3-VL 8B Thinking Alibaba58±1api $0.66/M +local~6.8GB2025-10
267Qwen3 32B Alibaba58±0api $0.13/M +local~20GB2025-04
more from Alibaba
18Qwen3.8 Max (2.4T-A95B) Alibaba78±1api $3.00/M +local~380GB2026-08
34Qwen3.8 Max (0902) Alibaba76±1api $3.00/M2026-09
37Qwen3.8-Flash-Next Alibaba75±1local~59GB2026-08

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #183 of 300 * · 2.1% of its score · 1 test

Beats o1 (2024-12-17), o3-mini, DeepSeek R1 0528 here, though they rank higher overall. Trails SmolLM3 3B, Jamba 1.5 Mini, Mellum2 12B-A2.5B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
BFCL v3 lab66.3% ✎█▄··▄█·█▄█▄█ beats 7/13medium · avg 59%▇ 2.1%blunt · at its level
coding▁▁▁▃▆██▇▆▄▃▁ #304 of 458 * · 4.6% of its score · 1 test

Beats Solar Open 100B (102B-A12B), GLM-4.7-Flash here, though they rank higher overall. Trails DeepSeek LLM 67B Chat, WizardLM 70B V1.0, Yi 1.5 9B Chat here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
LiveCodeBench v6 (window unspecified) lab42.6% ✎▃▅█▅▅▃▅···▃█ beats 6/16medium · avg 35%▇▇ 4.6%at its level
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #171 of 393 * · 2.8% of its score · 1 test

Beats MiniCPM5 2B, Seed-OSS 36B Instruct, GPT-5 nano here, though they rank higher overall. Trails LFM2.5 1.2B Instruct, LFM2.5 1.2B Thinking, Gemma 3 12B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
IFEval (aggregation unspecified) lab85.8% ✎▃··▃···██▅▃▃ beats 10/11easy · avg 77%▇ 2.8%typical
math▁▂▃▅▆▅▄█▄▁▁▁ #164 of 398 * · 10% of its score · 2 tests

Beats MiniMax-M2.7, MiniMax-M2.1, Mistral Medium 3.5 (128B) here, though they rank higher overall. Trails Olmo 3 7B Think, Nemotron Nano 9B v2, Qwen3-VL 4B Thinking here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
MathVista (testmini) lab80.1% ✎▃··▃▅·▃█▅▅▅█ beats 11/16medium · avg 69%▇▇ 5.4%typical
AIME 2025 lab69.3% ✎▃·▃▃█··▃▅▃·█ beats 8/12medium · avg 45%▇▇ 4.7%sharp · at its level
vision & documents▁▁▃▂▁▃▄▂▆▇█▄ #96 of 147 · 49% of its score · 13 tests

Beats DeepSeek V4 Flash Vision (Exp), Magistral Small 1.2 here, though they rank higher overall. Trails Llama 4 Maverick (402B-A17B), GPT-4o (2024-11-20), Qwen3-VL 8B Thinking here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
CharXiv (reasoning questions) lab48.9% ✎▄▂▄▆▄▄▆··▄█▂ beats 6/21medium · avg 39%▇▇▇▇ 10%at its level
MMMU (val) lab74.2% ✎▂···▆▂▂▂▄▄▄█ beats 12/16medium · avg 61%▇▇▇ 8.0%at its level
CharXiv descriptive questions lab85.5% ✎▄··▂▂▂·▂▄▄▂█ beats 7/14easy · avg 74%▇▇ 6.1%typical
MathVista (testmini) lab80.1% ✎▃··▃▅·▃█▅▅▅█ beats 11/16medium · avg 69%▇▇ 5.4%typical
MMStar lab72.1% ✎▂··▄▂▂·▄█▂▄▆ beats 10/16medium · avg 65%▇▇ 5.3%at its level
MMMU Pro lab60.4% ✎▃·█·▅▃▃▃▅▅▃█ beats 12/16medium · avg 46%▇▇ 4.7%at its level
HallusionBench lab61.5% ✎▃·▃·▅▃▅▅▃▅█▅ beats 10/16medium · avg 55%▇ 2.2%blunt · at its level
ChartQA lab86.8% ✎▂··▂·▂▃·▅▂▆█ beats 6/17easy · avg 85%▇ 1.5%blunt
5 more tests
testresultvs the others in its rundifficultycountswhy
RealWorldQA lab73.7% ✎▅▃·▃▅▅·█·▃▅█ beats 10/16easy · avg 70%▇ 1.5%blunt
DocVQA lab95.0% ✎▃··▅▃▅▃█▃█▅▅ beats 6/17very easy · avg 94%▇ 1.3%blunt · maxed out
ScreenSpot-Pro lab60.5% ✎▂····▂▄··▂█▆ beats 9/11medium · avg 46%▇ 1.3%blunt · at its level
CC-OCR lab80.7% ✎▃··▃▅·▅▅·█▅▃ beats 12/13easy · avg 74%▇ 1.0%blunt
OCRBench lab90.3% ✎▅█·▃▃█▃▅▅▃·▃ beats 15/16easy · avg 84%▇ 0.5%blunt
other tests8 not in a task · 37% of its score
testresultvs the others in its rundifficultycountswhy
GPQA (subset unspecified) lab70.4% ✎▂····▂·▂▂█▆█ beats 10/14hard · avg 49%▇▇▇▇ 9.3%sharp · at its level
MMLU-Pro lab77.8% ✎▄·▄·▄▄▄▄█▄██ beats 9/12medium · avg 63%▇▇▇ 6.9%at its level
MMLU-Redux lab88.4% ✎▃··▃·▃·█▃▅▃▅ beats 7/11easy · avg 79%▇▇ 5.4%typical
SuperGPQA lab53.1% ✎▄··▄·▄▄▄█▄██ beats 8/11medium · avg 37%▇▇ 4.6%at its level
AndroidWorld lab54.3% ✎▃··▃▅▃▃▃▃▃·█ beats 6/11medium · avg 43%▇ 3.3%blunt · at its level
LVBench lab62.5% ✎▅··▃·▅▃▃·█·▃ beats 7/10hard · avg 51%▇ 3.2%at its level
Video-MME (without subtitles) lab74.5% ✎▃··█··▅█▃▃▅▃ beats 10/13medium · avg 66%▇ 2.9%blunt · at its level
MVBench lab72.3% ✎▂▂▂·▂█··▆▄·▂ beats 9/13medium · avg 67%▇ 1.2%blunt · at its level

$ where to run it

Through an API

api · Alibaba Cloud Model Studio · $0.35/M · $0.2 in / $0.8 out per 1M tokens · 256k context · www.alibabacloud.com ↗ · price sourceUSD per 1M tokens, Alibaba Cloud Model Studio, Singapore (International deployment scope), API model qwen3-vl-30b-a3b-instruct (Non-Thinking mode only): input $0.2, output $0.8. Chinese-mainland/Global (US) deployments priced differently and not used, following the International convention of other Alibaba rows; free quota, batch and context-cache discounts excluded. Observed 2026-10-09.

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run Qwen3-VL 30B-A3B Instruct · needs ~18.6 GB · from huggingface.co/Qwen (official)
    chat in the terminal
    llama cli -hf Qwen/Qwen3-VL-30B-A3B-Instruct-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf Qwen/Qwen3-VL-30B-A3B-Instruct-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run Qwen3-VL 30B-A3B Instruct · needs ~18.6 GB · from huggingface.co/Qwen (official)

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/Qwen/Qwen3-VL-30B-A3B-Instruct-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run Qwen3-VL 30B-A3B Instruct · needs ~20 GB · from ollama.com
    chat in the terminal
    ollama run qwen3-vl:30b-a3b-instruct

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Qwen3-VL 30B-A3B Instruct · needs ~18.3 GB, Mac only · from huggingface.co/lmstudio-community
    chat in the terminal
    mlx_lm.chat --model lmstudio-community/Qwen3-VL-30B-A3B-Instruct-MLX-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model lmstudio-community/Qwen3-VL-30B-A3B-Instruct-MLX-4bit

Official weights and documentation: huggingface.co/Qwen ↗

$ how the score was worked out · Qwen (Alibaba) vendor-reporte… 100%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
Qwen (Alibaba) vendor-reported benchmark scores for its own models (qwen.ai launch posts, Qwen2.5-VL and Qwen3-VL technical reports, Qwen3.8-Flash-Next card) · lab-reported25▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 100%Factual vendor-reported score (not licensed content)
prior · pull toward the middle▇ 0%
weighted average= 57.8

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json