Qwen2-VL-7B-Instruct
Alibaba · Qwen/Qwen2-VL-7B-Instruct
Weights license: Apache 2.0 · code and components may use other terms
Quality
Quality · LMArena · vision
LMArena 2026-10-19 · human-preference rating · higher is better · this model outlined
- Color = publisher
Source: LMArena 2026-10-19 · collected 2026-10-25 · dataset · CC BY 4.0 · models without a rating are not drawn (no rating ≠ 0)
1,031 ± 11 · LMArena · vision · Confidence interval 1,020–1,041 · 5,766 votes
Reported by the publisher
- MMMU val
- 54.1Image Benchmarks
- DocVQA test
- 94.5Image Benchmarks
- InfoVQA test
- 76.5Image Benchmarks
- ChartQA test
- 83.0Image Benchmarks
- TextVQA val
- 84.3Image Benchmarks
- OCRBench
- 845Image Benchmarks
- MTVQA
- 26.3Image Benchmarks
- VCR en easy
- 89.70Image Benchmarks
Every reported result
- VCR zh easy
- 59.94Image Benchmarks
- RealWorldQA
- 70.1Image Benchmarks
- MME sum
- 2,326.8Image Benchmarks
- MMBench-EN test
- 83.0Image Benchmarks
- MMBench-CN test
- 80.5Image Benchmarks
- MMBench-V1.1 test
- 80.7Image Benchmarks
- MMT-Bench test
- 63.7Image Benchmarks
- MMStar
- 60.7Image Benchmarks
- MMVet GPT-4-Turbo
- 62.0Image Benchmarks
- HallBench avg
- 50.6Image Benchmarks
- MathVista testmini
- 58.2Image Benchmarks
- MathVision
- 16.3Image Benchmarks
- MVBench
- 67.0Video Benchmarks
- PerceptionTest test
- 62.3Video Benchmarks
- EgoSchema test
- 66.7Video Benchmarks
Values as printed on the card, with the publisher's own settings (card section: Evaluation › Image Benchmarks; Evaluation › Video Benchmarks); column “Qwen2-VL-7B”. Not comparable across publishers and not used in any ranking. Model card
Quality × memory
No comparable quality × memory pair
This entry has no score on a benchmark that is drawn against planning memory.
Alibaba released Qwen2-VL-7B-Instruct on 2024-08-30: a vision-language model that answers about images, with 8.29B parameters on the Hugging Face count and the Apache 2.0 license. Languages the publisher declares: en.
License: commercial use allowed. At Q4 the weights load about 6.2 GB before each machine's reserve; image resolution, video frames and context add memory, and the vision encoder needs support in the runtime you pick.
Runs on
Fast memory holds it 90
The planning estimate fits the memory the model can use quickly: graphics card VRAM, or unified memory within the macOS GPU share.
Show all 90
Within memory, outside the fast part 26
Within the machine's memory, outside its fast part: on a graphics card the layers that do not fit run on the CPU from PC RAM (the listed RAM, or 32 GB assumed); on a Mac, above the macOS default GPU share. Not a speed figure.
Show all 26
One memory tier short 3
The same machines with the next memory size up would hold it in fast memory.
Hardware profiles grouped by the Q4 planning estimate on each machine (dedicated GPUs reserve 2 GB; unified and system memory 25%, 2 to 8 GB): the fast memory holds it, it is within memory but outside the fast part (layers offloaded to PC RAM, or above the macOS GPU share), or the machine is one memory tier short. Not a measured fit or speed.
Measured speed
No published measurement for this checkpoint
Tokens per second are shown only when a source publishes them with hardware, build and setup. They are never estimated.
How to run
Runtime compatibility and tokens per second have not been measured for this entry. Published metadata and a memory estimate do not prove a successful local run.
Library in source: transformers
Documented ways to run it
Copied as printed in the linked source (read 2026-10-18); not run by this site.
pip install qwen-vl-utilsfrom transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
from qwen_vl_utils import process_vision_info
# default: Load the model on the available device(s)
model = Qwen2VLForConditionalGeneration.from_pretrained(
"Qwen/Qwen2-VL-7B-Instruct", torch_dtype="auto", device_map="auto"
)
# We recommend enabling flash_attention_2 for better acceleration and memory saving, especially in multi-image and video scenarios.
# model = Qwen2VLForConditionalGeneration.from_pretrained(
# "Qwen/Qwen2-VL-7B-Instruct",
# torch_dtype=torch.bfloat16,
# attn_implementation="flash_attention_2",
# device_map="auto",
# )
# default processer
processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")
# The default range for the number of visual tokens per image in the model is 4-16384. You can set min_pixels and max_pixels according to your needs, such as a token count range of 256-1280, to balance speed and memory usage.
# min_pixels = 256*28*28
# max_pixels = 1280*28*28
# processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", min_pixels=min_pixels, max_pixels=max_pixels)
messages = [
{
"role": "user",
"content": [
{
"type": "image",
"image": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg",
},
{"type": "text", "text": "Describe this image."},
],
}
]
# Preparation for inference
text = processor.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
image_inputs, video_inputs = process_vision_info(messages)
inputs = processor(
text=[text],
images=image_inputs,
videos=video_inputs,
padding=True,
return_tensors="pt",
)
inputs = inputs.to("cuda")
# Inference: Generation of the output
generated_ids = model.generate(**inputs, max_new_tokens=128)
generated_ids_trimmed = [
out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]
output_text = processor.batch_decode(
generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
print(output_text)Technical details
What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.
- Checkpoint / source
- Qwen/Qwen2-VL-7B-Instruct
- Task
- image-text-to-text
- Architecture
- Qwen2VLForConditionalGeneration
- Library in source
- transformers
- Weight license
- Apache 2.0Qwen/Qwen2-VL-7B-Instruct model card (Hugging Face) · read 2026-10-18source
- Commercial use
- Allowed Apache License 2.0 text (SPDX) · read 2026-10-18license text
- Languages
- enModel card metadata · read 2026-10-25source
- Released
- Qwen3-VL README news (GitHub) · read 2026-10-18source
Every stored field
- Revision checked
- eed13092ef92e448dd6875b2a00151bd3f7db0ac
- Architecture
- Qwen2VLForConditionalGeneration
- Library
- transformers
- Hub task tag
- image-text-to-text
- Declared base models
- Qwen/Qwen2-VL-7B
- Formats
- transformers, safetensors
- Access
- Open download
- Publisher kind
- Official
- Hub parameter count
- 8,291,375,616safetensors.total
- Repository created
- 2024-08-28creation is not the model's release
- Repository updated
- 2025-02-06a commit is not a release date
- Hub downloads (30 days) · likes
- 601,271 · 1,295this repository only · read 2026-10-25
- Catalog state
- Hub config or weights found; no inventory benchmark link
Sources
Identity and published specifications come from the checkpoint page. Benchmark scores belong to the named release and do not measure speed on your computer.
- Publisher model card
- License: Qwen/Qwen2-VL-7B-Instruct model card (Hugging Face)
- Release date: Qwen3-VL README news (GitHub)
Curated facts checked: 2026-10-18. The license and release date cite the publisher's sources; parameters, files and task come from the Hugging Face record, and memory is a labelled estimate from its parameter count. A value no source states stays empty; a repository commit is not a model release date.






























































