whatmodel.sh / Qwen3-VL 8B Instruct

Qwen3-VL 8B Instruct

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

Qwen3-VL 8B Instruct is an open-weights language model from Alibaba, released October 2025. Its published size is 8.77B parameters and a 256k-token context window. Licence: apache-2.0.

Catalogue identity notes

Maker repo Qwen/Qwen3-VL-8B-Instruct @0c351dd (licence tag apache-2.0, created 2025-10-11); the separate Instruct checkpoint, not Qwen3-VL-8B-Thinking. Scored from the Qwen3-VL Technical Report (arXiv 2511.21631v2, Tables 3/4 and 7-10: column 'Qwen3-VL 8B instruct'), whose instruct and thinking columns are the two published checkpoints. Released 2025-10: month the maker's Hugging Face repo was created.

Model information sources

Release date ↗ · Weights licence ↗

8,767,123,696 parameters in the Hugging Face safetensors metadata of Qwen/Qwen3-VL-8B-Instruct @0c351dd (vision encoder included); dense, so active = total.

55.1±0.7#265 of 719 LLMsfrom 31 benchmark results across 5 sources

$ how it compares

The LLMs ranked either side of Qwen3-VL 8B Instruct, out of 828 with a score, then more from Alibaba.

#modelscorewhereRAMreleased
317Qwen3 4B Instruct 2507 Alibaba55±1local~3.9GB2025-08
318Qwen2.5 Coder 14B Instruct Alibaba55±1local~10GB2024-11
319Qwen3-VL 8B Instruct Alibaba55±1api $0.31/M +local~6.8GB2025-10
320Spark-X2.5 1.7B iFLYTEK (SparkLLM)55±1local~2.7GB2026-08
321GigaChat 3.5 Ultra Instruct (432B-A28B) Sber55±1local~457GB2026-07
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▁▁▃▆█▆▅▅▅▄▃▁ #192 of 300 * · 2.9% of its score · 2 tests

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 6/13medium · avg 59%▇ 1.9%blunt · at its level
BFCL (version not stated)66.2% ✎▄▆▄▂·█▆·▄··▂ beats 6/17easy · avg 66%▇ 1.0%blunt · at its level
coding▁▁▁▃▆██▇▆▄▃▁ #325 of 458 * · 4.1% of its score · 1 test

Beats GLM-4.7-Flash here, though they rank higher overall. Trails Phi-3-mini-128k-instruct (Apr 2024), DeepSeek LLM 67B Chat, WizardLM 70B V1.0 here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
LiveCodeBench v6 (window unspecified) lab39.3% ✎▃▅█▅▅▃▅···▃█ beats 5/16medium · avg 35%▇▇ 4.1%at its level
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #188 of 393 * · 2.7% of its score · 1 test

Beats Seed-OSS 36B Instruct, Granite 4.2 8B, Qwen3-VL 30B-A3B Thinking 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) lab83.7% ✎▃··▃···██▅▃▃ beats 7/11easy · avg 77%▇ 2.7%typical
math▁▂▃▅▆▅▄█▄▁▁▁ #184 of 398 · 12% of its score · 4 tests

Beats MiniMax-M2.1, DeepSeek V3 0324, Qwen3 235B-A22B (Apr 2025, hybrid) here, though they rank higher overall. Trails Phi-4-mini-reasoning 3.8B, OpenThinker3 7B, Olmo 3 7B Think here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
MathVista (testmini) lab77.2% ✎▃··▃▅·▃█▅▅▅█ beats 9/16medium · avg 69%▇▇ 5.2%at its level
AIME 2025 lab45.9% ✎▃·▃▃█··▃▅▃·█ beats 3/12medium · avg 45%▇▇ 4.5%sharp · at its level
OMEGA (math generalisation)32.3% ✎█···▄·▄·▄··█ beats 6/6hard · avg 20%▇ 1.9%at its level
MATH (subset unspecified)91.6% ✎██····██·███ beats 6/6easy · avg 71%▇ 0.8%noisy
vision & documents▁▁▃▂▁▃▄▂▆▇█▄ #104 of 147 · 49% of its score · 16 tests

Beats DeepSeek V4 Flash Vision (Exp), GPT-4o (2024-08-06), GPT-4 Turbo (2024-04-09) here, though they rank higher overall. Trails Llama 4 Scout (109B-A17B) here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
CharXiv (reasoning questions) lab46.4% ✎▄▂▄▆▄▄▆··▄█▂ beats 5/21medium · avg 39%▇▇▇ 8.7%at its level
MMMU (val) lab69.6% ✎▂···▆▂▂▂▄▄▄█ beats 8/16medium · avg 61%▇▇▇ 7.4%at its level
CharXiv descriptive questions lab83.0% ✎▄··▂▂▂·▃▄▄▂█ beats 5/14easy · avg 74%▇▇ 6.2%typical
MathVista (testmini) lab77.2% ✎▃··▃▅·▃█▅▅▅█ beats 9/16medium · avg 69%▇▇ 5.2%at its level
MMStar lab70.9% ✎▂··▄▂▂·▄█▂▄▆ beats 9/16medium · avg 65%▇▇ 4.8%at its level
MMMU Pro lab55.9% ✎▃·█·▅▃▃▃▅▅▃█ beats 9/16medium · avg 46%▇▇ 4.2%at its level
ChartArena chart parsing (English, mAP_high average)37.3% ✎▁·▁·▃▃▁▇▃█·▄ beats 11/22hard · avg 27%▇ 2.3%blunt · at its level
HallusionBench lab61.1% ✎▃·▃·▅▃▅▅▃▅█▅ beats 9/16medium · avg 55%▇ 2.0%blunt · at its level
8 more tests
testresultvs the others in its rundifficultycountswhy
BEAR-Bench (bilingual RU/EN enterprise and academic document images, overall accuracy)25.4% ✎▃▃·▅·▃·█·█▅▅ beats 3/14hard · avg 24%▇ 1.6%at its level
ChartQA lab89.6% ✎▂··▂·▂▃·▅▂▆█ beats 15/17easy · avg 85%▇ 1.5%blunt
RealWorldQA lab71.5% ✎▅▃·▃▅▅·█·▃▅█ beats 7/16easy · avg 70%▇ 1.4%blunt · at its level
DocVQA lab96.1% ✎▃··▅▃▅▃█▃█▅▅ beats 11/17very easy · avg 94%▇ 1.3%blunt · maxed out
ScreenSpot-Pro lab54.6% ✎▂····▂▄··▃█▆ beats 4/11medium · avg 46%▇ 1.2%blunt · at its level
CC-OCR lab79.9% ✎▃··▃▅·▅▅·█▅▃ beats 10/13easy · avg 74%▇ 0.9%blunt
OCRBench lab89.6% ✎▅█·▃▃█▃▅▅▃·▃ beats 14/16easy · avg 84%▇ 0.5%blunt
PureDocBench document parsing (Avg3 over Clean/Digital/Real)69.1% ✎▂·▁·▂▂▅█·▁▂▂ beats 11/22medium · avg 62%▇ 0.1%blunt · at its level
other tests8 not in a task · 34% of its score
testresultvs the others in its rundifficultycountswhy
GPQA (subset unspecified) lab61.9% ✎▂····▂·▂▂█▆█ beats 4/14hard · avg 49%▇▇▇ 8.2%sharp · at its level
MMLU-Pro lab71.6% ✎▄·▄·▄▄▄▄█▄██ beats 5/12medium · avg 63%▇▇▇ 6.7%at its level
MMLU-Redux lab84.9% ✎▃··▃·▃·█▃▅▃▅ beats 4/11easy · avg 79%▇▇ 5.5%typical
SuperGPQA lab44.5% ✎▄··▄·▄▄▄█▄██ beats 4/11medium · avg 37%▇▇ 4.0%at its level
AndroidWorld lab47.6% ✎▃··▃▅▃▃▃▃▃·█ beats 3/11medium · avg 43%▇ 2.9%blunt · at its level
LVBench lab58.0% ✎▅··▃·▅▃▃·█·▃ beats 5/10hard · avg 51%▇ 2.8%at its level
Video-MME (without subtitles) lab71.4% ✎▃··█··▅█▃▃▅▃ beats 7/13medium · avg 66%▇ 2.6%blunt · at its level
MVBench lab68.7% ✎▂▂▂·▂█··▆▄·▂ beats 4/13medium · avg 67%▇ 1.1%blunt · at its level

$ where to run it

Through an API

api · Alibaba Cloud Model Studio · $0.31/M · $0.18 in / $0.7 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-8b-instruct (Non-Thinking mode only): input $0.18, output $0.7. 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 8B Instruct · needs ~5 GB · from huggingface.co/Qwen (official)
    chat in the terminal
    llama cli -hf Qwen/Qwen3-VL-8B-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-8B-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 8B Instruct · needs ~5 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-8B-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 8B Instruct · needs ~6.1 GB · from ollama.com
    chat in the terminal
    ollama run qwen3-vl:8b-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 8B Instruct · needs ~5.8 GB, Mac only · from huggingface.co/lmstudio-community
    chat in the terminal
    mlx_lm.chat --model lmstudio-community/Qwen3-VL-8B-Instruct-MLX-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model lmstudio-community/Qwen3-VL-8B-Instruct-MLX-4bit

Official weights and documentation: huggingface.co/Qwen ↗

$ how the score was worked out · Qwen (Alibaba) vendor-reporte… 92%, allenai/Olmo-3-7B-Instruct mo… 4%, ChartArena: Benchmarking Char… 2% and 2 more

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▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 92%Factual vendor-reported score (not licensed content)
allenai/Olmo-3-7B-Instruct model card3▇ 4%Apache-2.0
ChartArena: Benchmarking Chart Parsing across Languages, Scenarios, and Formats (arXiv:2606.01348v3)1▇ 2%CC BY 4.0
BEAR-Bench: A Bilingual Enterprise and Academic Reasoning Benchmark for Multimodal Models (arXiv:2608.17895v1)1▇ 2%CC BY 4.0
How Far Is Document Parsing from Solved? PureDocBench: A Source-Traceable Benchmark across Clean, Degraded, and Real-World Settings (arXiv:2605.07492v2)1▇ 0%CC BY 4.0
prior · pull toward the middle▇ 0%
weighted average= 55.1

$ 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