HO

holo

HoloGGUF by gguf-orgUpdated 2025-09-22
Download on Hugging Face
Parameters4.07B
Context—
Layers—
Architecturepig
Smallest quant3.5 GB
Quick answer

holo needs about 5.1 GB of memory at Q4_K_M (a 4.1 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 16 quants, from 3.5 GB (Q2_K) to 8.1 GB (BF16).

4.07B parameter model for computer‑use agents

Holo is a 4.07B parameter model fine‑tuned for UI localization and UI‑based question answering. It is part of the Holo1.5 family and supports high‑resolution interfaces up to 3840×2160 pixels. The model is available in several GGUF quantizations for different memory budgets.

Good at

  • UI localization
  • UI‑based question answering
  • High‑resolution interface understanding

Limits

  • No vision support beyond UI images
  • License is other, not fully open
  • No explicit context length specified

Best for

  • Building computer‑use agents
  • Automating web and desktop tasks
  • UI element detection

Who it’s for. Developers building computer‑use agents who need reliable UI understanding. The model is suitable for research and commercial use under the provided license terms.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K3.5 GB 4.4 GB
Fits Get file →
IQ3_XXS3.6 GB 4.6 GB
Fits Get file →
IQ3_S3.8 GB 4.7 GB
Fits Get file →
Q3_K_M3.8 GB 4.7 GB
Fits Get file →
IQ4_XS4.1 GB 5.0 GB
Fits Get file →
IQ4_NL4.1 GB 5.1 GB
Fits Get file →
Q4_04.1 GB 5.1 GB
Fits Get file →
Q4_K_M4.1 GB 5.1 GB
Fits Get file →
Q4_14.3 GB 5.3 GB
Fits Get file →
Q5_04.5 GB 5.5 GB
Fits Get file →
Q5_K_M4.5 GB 5.5 GB
Fits Get file →
Q5_14.7 GB 5.7 GB
Fits Get file →
Q6_K4.9 GB 5.9 GB
Fits Get file →
Q8_05.5 GB 6.7 GB
Fits Get file →
F168.1 GB 9.6 GB
Too big Get file →
BF168.1 GB 9.6 GB
Too big Get file →

For 8 GB VRAM, use Q2_K (3.5 GB file, 4.4 GB memory) or IQ3_XXS (3.65 GB file, 4.6 GB memory). For 12 GB VRAM, Q4_K_M (4.15 GB file, 5.1 GB memory) or Q5_0 (4.49 GB file, 5.5 GB memory) are good choices. Higher precision quants like Q8_0 or F16 require more than 16 GB VRAM.

What GPU do you need for holo?

These figures add the file size, an fp16 KV cache at an 8K context window 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: no quant fits at an 8K context.
  • 6 GB VRAM: the best quant that fits is Q6_K (4.9 GB file, about 5.9 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (5.5 GB file, about 6.7 GB in use).
  • 12 GB VRAM: the best quant that fits is BF16 (8.1 GB file, about 9.6 GB in use).

Which holo quant should you choose?

Q4_K_M is the usual starting point for holo: it balances file size and quality well. Here it is a 4.1 GB file.

The Q8_0 file is 5.5 GB, about 1.3 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 Q2_K at 3.5 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturepig
Parameters4.07B

Source

Quantized bygguf-org
FamilyHolo
Licenseother
Downloads (HF)930
Last updated2025-09-22

Frequently asked questions about holo

How much VRAM does holo need?

About 5.1 GB at Q4_K_M with an 8K context window. That is the 4.1 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 holo on an 8 GB GPU?

Yes, with Q8_0: the 5.5 GB file needs about 6.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of holo GGUF?

It depends on the quant: Q2_K is 3.5 GB, IQ3_XXS is 3.6 GB, IQ3_S is 3.8 GB, Q3_K_M is 3.8 GB, IQ4_XS is 4.1 GB, IQ4_NL is 4.1 GB and 10 more.

Where can I download holo GGUF?

The files are on Hugging Face in the gguf-org/holo-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 holo use?

Hugging Face lists the license as other. 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 holo on 8 GB VRAM?

Yes, use Q2_K or IQ3_XXS quantizations; they require about 4.4–4.6 GB memory at 8 k context.

Does holo support vision tasks?

It is designed for UI images, not general vision tasks. Use it for interface understanding only.

What license applies to holo?

The model is released under an other license; check the repository for specific terms.

Other GGUFs of Holo1.5-3B

Similar models

Browse all →

More from gguf-org

All by gguf-org →

More Holo models

All Holo →

Other small models

All small models →