ID

ideogram 4

IdeogramGGUF by molbalUpdated 2026-06-15
Download on Hugging Face
Parameters9.28B
Context—
Layers—
Architecture—
Smallest quant5.6 GB
Quick answer

ideogram 4 needs about 8.1 GB of memory at Q5_0 (a 6.8 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 10 quants, from 5.6 GB (Q4_0) to 10.1 GB (Q8_0).

Ideogram 4 diffusion transformer in GGUF format

Ideogram 4 is a 9.8 billion‑parameter diffusion transformer converted to GGUF for use with ComfyUI. It supports text‑guided image generation via separate main and unconditional transformer files.

Good at

  • Text‑guided diffusion
  • ComfyUI integration
  • Multiple quant levels

Limits

  • Requires separate main and unconditional files
  • No standalone package
  • No vision encoder included

Best for

  • ComfyUI workflows needing diffusion models
  • Image generation with CFG
  • Custom quant pairing

Who it’s for. Users of ComfyUI who need a diffusion transformer for Ideogram 4 workflows. The model is provided as GGUF files and must be paired with the required encoder and VAE assets.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_05.6 GB 6.8 GB
Fits Get file →
Q4_05.6 GB 6.8 GB
Fits Get file →
Q4_16.2 GB 7.4 GB
Tight Get file →
Q4_16.2 GB 7.4 GB
Tight Get file →
Q5_06.8 GB 8.1 GB
Too big Get file →
Q5_06.8 GB 8.1 GB
Too big Get file →
Q5_17.3 GB 8.7 GB
Too big Get file →
Q5_17.3 GB 8.7 GB
Too big Get file →
Q8_010.1 GB 11.9 GB
Too big Get file →
Q8_010.1 GB 11.9 GB
Too big Get file →

For 8 GB VRAM, use Q4_0 (5.64 GB) or Q4_1 (6.21 GB) files; memory estimates are 6.8 GB and 7.4 GB at 8k context. For 12 GB VRAM, Q5_0 (6.77 GB) or Q5_1 (7.33 GB) are suitable, estimating 8.1 GB and 8.7 GB. Q8_0 (10.14 GB) requires 11.9 GB VRAM.

What GPU do you need for ideogram 4?

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: no quant fits at an 8K context.
  • 8 GB VRAM: the best quant that fits is Q4_1 (6.2 GB file, about 7.4 GB in use).
  • 12 GB VRAM: the best quant that fits is Q8_0 (10.1 GB file, about 11.9 GB in use).

Which ideogram 4 quant should you choose?

Q5_0 (6.8 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

If memory is tight, the smallest file is Q4_0 at 5.6 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Parameters9.28B

Source

Quantized bymolbal
FamilyIdeogram
Licenseother
Downloads (HF)2,140
Last updated2026-06-15

Frequently asked questions about ideogram 4

How much VRAM does ideogram 4 need?

About 8.1 GB at Q5_0 with an 8K context window. That is the 6.8 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 ideogram 4 on an 8 GB GPU?

Yes, with Q4_1: the 6.2 GB file needs about 7.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of ideogram 4 GGUF?

It depends on the quant: Q4_0 is 5.6 GB, Q4_0 is 5.6 GB, Q4_1 is 6.2 GB, Q4_1 is 6.2 GB, Q5_0 is 6.8 GB, Q5_0 is 6.8 GB and 4 more.

Where can I download ideogram 4 GGUF?

The files are on Hugging Face in the molbal/ideogram-4-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 ideogram 4 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.

Do I need additional assets to use this model?

Yes, the model requires the text or multimodal encoder and VAE used by Ideogram 4 in ComfyUI.

Is the model a mixture of experts?

No, Ideogram 4 is a standard transformer, not a mixture of experts.

Other GGUFs of ideogram-4-fp8

Similar models

Browse all →

More from molbal

All by molbal →

More Ideogram models

All Ideogram →

Other small models

All small models →