Skip to content
Local AI Wizard
Your machine
Theme

Ideogram 4.0

Ideogram · collections/ideogram-ai/ideogram-4

Weights license: Publisher terms: Ideogram 4 Non-Commercial · code and components may use other terms

QualityArtificial Analysis score, rank and sample count excluded: its free tier is internal use only. Source checked 2026-09-22.No comparable score
Published repository size16.1 GBAll published files · not runtime memory
Parameters9.3 Bmanufacturer model card
Published components3Listed in the model card

Quality

No comparable score

Artificial Analysis score, rank and sample count excluded: its free tier is internal use only. Source checked 2026-09-22. Leaderboard (link only)

Reported by the publisher

This model card has not been read for benchmark results yet. card not read yet

Runs on

Enough memory 22

A publisher documents a path within this memory, on this platform (CUDA or Apple MPS). Conditional paths need offload or quantization.

Show all 22

One memory tier short 16

The same machines with the next memory size up would have enough.

Show all 16

Hardware profiles grouped by the publisher's documented minimum memory path and platform. Not a measured run.

Plan memory for this model

Consider it for

Consider Ideogram 4.0 from Ideogram for the text-to-image task (9.3B parameters). The publisher's announcement states that the official NF4 checkpoint fits on one 24 GB CUDA GPU (not re-verified on 2026-10-17: the announcement could not be opened).

Check first

This is the NF4 path, not an FP8 or Fast/Instant requirement. Resolution and runtime configuration can affect memory use.

Measured speed

No published measurement for this checkpoint

Generation time per image is shown only when a source publishes it with hardware, build and setup. It is never estimated.

How to run

The publisher's announcement states that the official NF4 checkpoint fits on one 24 GB CUDA GPU (not re-verified on 2026-10-17: the announcement could not be opened). This is the NF4 path, not an FP8 or Fast/Instant requirement. Resolution and runtime configuration can affect memory use. Official hardware evidence

Documented ways to run it

The model card could not be read (missing or gated)

Evidence level: Publisher documentation linked; no execution reproduced by this site.

Before downloading
  1. Check the exact checkpoint and license linked here.
  2. Confirm a compatible runtime and precision or quantization for your platform.
  3. Allow for context, batch size and any extra media components beyond the weights.

Technical details

What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.

Task and modality
text-to-image · image
Count and basis
9.3B · manufacturer model card
Released
Ideogram 4.0 Technical Details: Open model at the forefront of design (Ideogram blog) · read 2026-10-09source
Published repository size
16.1 GBNot runtime memory or a minimum download.files
Published components
9.3B diffusion transformer · Qwen3-VL-8B text encoder · NF4 checkpoint
Runtime and precision
No local execution verified.
Commercial use
Not allowed Ideogram 4 Non-Commercial license (ideogram-oss/ideogram-4) · read 2026-10-18license text

The source labels do not establish an exact sampler preset mapping. The 24 GB path and 16.1 GB repository figure apply to NF4. The leaderboard rows “Ideogram 4.0” and “Ideogram 4.0 (Quality)” name the same open model and share this page. “Quality” is one of the publisher's three sampler presets for the same weights (V4_QUALITY_48, 48 steps); no readable primary source maps either leaderboard row to a specific preset, or says whether it ran the open NF4 weights or the hosted model. The memory figures here are for the open NF4 checkpoint. Two of them could not be re-read in the 2026-10-17 review and are unverified: the 24 GB single-GPU figure cites the ideogram.ai announcement, which could not be opened (a secondary article repeats it), and the 16.1 GB repository size comes from a gated repository that returned access denied. The 9.3B parameter count and the sampler presets were confirmed in the publisher's repository.

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.

Catalog reviewed: 2026-09-22.