WE

WeVisOne VL

WevisoneGGUF by PALLER-NOVUSUpdated 2026-10-26
Download on Hugging Face
Parameters407M
Context256K
Layers28
Architectureqwen3vl
Smallest quant0.8 GB
Quick answer

WeVisOne VL needs about 2.7 GB of memory at Q4_K_M (a 1.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 18 quants, from 0.8 GB to 8.8 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 3.8 GB.

WeVisOne VL – a 0.407B vision‑language model for structured perception

WeVisOne VL is a 0.407B parameter model built on Qwen3‑VL. It supports object‑level tasks such as detection, referring, counting, pointing, OCR and pixel‑level tasks like segmentation and alpha matting. The model is released under Apache‑2.0 and is available in multiple GGUF quantizations.

Good at

  • Object detection and referring localization
  • Pixel‑level segmentation and alpha matting
  • OCR localization

Limits

  • No vision support beyond structured tasks
  • Requires 8k context for full performance
  • No autonomous decision use

Best for

  • Research in visual grounding
  • Image matting experiments
  • OCR localization pipelines

Who it’s for. Researchers and developers needing a lightweight vision‑language model for structured perception tasks. The model is suitable for local inference on GPUs with at least 8 GB VRAM and can be run with CPU offload for larger models.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.8 GB 2.3 GB
Fits Get file →
—0.8 GB 2.3 GB
Fits Get file →
Q2_K0.9 GB 2.3 GB
Fits Get file →
Q3_K_M1.1 GB 2.5 GB
Fits Get file →
IQ4_XS1.2 GB 2.6 GB
Fits Get file →
Q4_K_M1.3 GB 2.7 GB
Fits Get file →
Q5_K_M1.5 GB 2.9 GB
Fits Get file →
Q6_K1.7 GB 3.1 GB
Fits Get file →
Q2_K1.8 GB 3.2 GB
Fits Get file →
Q8_02.2 GB 3.6 GB
Fits Get file →
Q3_K_M2.2 GB 3.7 GB
Fits Get file →
IQ4_XS2.5 GB 3.9 GB
Fits Get file →
Q4_K_M2.7 GB 4.2 GB
Fits Get file →
Q5_K_M3.2 GB 4.6 GB
Fits Get file →
Q6_K3.6 GB 5.1 GB
Fits Get file →
F164.1 GB 5.5 GB
Fits Get file →
Q8_04.7 GB 6.1 GB
Fits Get file →
F168.8 GB 10.3 GB
Too big Get file →

For 8 GB VRAM use Q4_K_M (file size 1.28 GB, estimated 2.7 GB memory at 8k context). For 12 GB VRAM Q5_K_M (1.47 GB, 2.9 GB memory) or Q6_K (1.68 GB, 3.1 GB). For 16 GB VRAM Q8_0 (2.17 GB, 3.6 GB) or F16 (4.08 GB, 5.5 GB). Estimates are approximate.

What GPU do you need for WeVisOne VL?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is IQ4_XS (2.5 GB file, about 3.9 GB in use).
  • 6 GB VRAM: the best quant that fits is F16 (4.1 GB file, about 5.5 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (4.7 GB file, about 6.1 GB in use).
  • 12 GB VRAM: the best quant that fits is F16 (8.8 GB file, about 10.3 GB in use).

Which WeVisOne VL quant should you choose?

Q4_K_M is the usual starting point for WeVisOne VL: it balances file size and quality well. Here it is a 1.3 GB file.

The Q8_0 file is 2.2 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is the first one in the table at 0.8 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen3vl
Parameters407M
Layers28
Trained context262,144 tokens
KV cache (fp16)112 KB per token

Source

Quantized byPALLER-NOVUS
FamilyWevisone
Licenseapache-2.0
Downloads (HF)1,511
Last updated2026-10-26

Frequently asked questions about WeVisOne VL

How much VRAM does WeVisOne VL need?

About 2.7 GB at Q4_K_M with an 8K context window. That is the 1.3 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run WeVisOne VL on an 8 GB GPU?

Yes, with Q8_0: the 4.7 GB file needs about 6.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of WeVisOne VL GGUF?

It depends on the quant: file is 0.8 GB, file is 0.8 GB, Q2_K is 0.9 GB, Q3_K_M is 1.1 GB, IQ4_XS is 1.2 GB, Q4_K_M is 1.3 GB and 12 more.

What context length does WeVisOne VL support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.8 GB at 32K.

Where can I download WeVisOne VL GGUF?

The files are on Hugging Face in the PALLER-NOVUS/WeVisOne-VL-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does WeVisOne VL use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run WeVisOne VL on 8 GB VRAM?

Yes, the Q4_K_M quantization requires about 2.7 GB memory at 8k context, leaving headroom for buffers.

Does the model support image segmentation?

Yes, it includes a pixel‑decoder for referring segmentation and alpha matting.

Is the model licensed for commercial use?

It is released under the Apache‑2.0 license, which permits commercial use.

Other GGUFs of WeVisOne-VL-2B

Similar models

Browse all →

More from PALLER-NOVUS

All by PALLER-NOVUS →

More Wevisone models

All Wevisone →

Other tiny models

All tiny models →