Skip to content
localmodel.run

Text model · Qwen2.5-VL

Qwen2.5-VL 7B requirements

Qwen2.5-VL family · 8.29B params · released Jan 2025. Minimum to run at Q4_K_M: Nvidia GeForce RTX 3060 (12GB).

Vision · accepts imagesLicenseApache-2.0· Commercial OK↓ 4.9M/mo♥ 1.6Kon HuggingFace
Q4_K_M
5.62 GB
Q8_0
9.59 GB
Total @ Q4 (4k)
~7.1 GB
Context
32 k

Quantization sizes

GGUF quantson disk
QuantizationSize on disk
Q2_K3.5 GB est
Q3_K_M4.1 GB est
Q4_K_M (default)5.62 GB
Q5_K_M5.9 GB est
Q6_K6.8 GB est
Q8_09.59 GB
FP1616.6 GB est

Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.

Run it

llama.cpp
$ llama-cli -hf ggml-org/Qwen2.5-VL-7B-Instruct-GGUF:Q4_K_M
LM Studio
$ lms get ggml-org/Qwen2.5-VL-7B-Instruct-GGUF

Which devices can run Qwen2.5-VL 7B?

FAQ

How much VRAM or RAM does Qwen2.5-VL 7B need?

At Q4_K_M, Qwen2.5-VL 7B needs about 7.1 GB (weights ~5.62 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~11.1 GB.

Can Qwen2.5-VL 7B run on a laptop?

Qwen2.5-VL 7B is large; you need a 24 GB+ GPU or a 32-48 GB Mac at Q4_K_M.

Can I use Qwen2.5-VL 7B commercially?

Yes. Qwen2.5-VL 7B is licensed Apache-2.0, which permits commercial use.

Qwen2.5-VL 7B vision-language model: chats about images. Q4_K_M ~4.4GB plus a ~1.26GB fp16 vision projector (mmproj), folded into the totals. Sizes from the ggml-org GGUF repo. Vision needs an mmproj-capable runtime; many mobile runtimes do not support it yet.

Sources

Memory figures are estimates. See methodology.