Model reference · open weights
ideogram-4 is an open-weight image model from unsloth. ideogram-4-fp8 (FP8) weighs 27.5 GB; the smallest configuration that runs it is RTX 4060 Ti 16 GB.
For self-hosting, compare this build's requirements below with the GPU you plan to rent.
Summary of the unsloth/ideogram-4-fp8 model card, 2026-10-26
What it is
| Released by | unsloth |
|---|---|
| Released | 2026-08-05 |
| Parameters | 9.3B |
| VRAM | 27.5 GB for the weights |
What it runs on
| Card | One image | Memory |
|---|---|---|
| RTX 3060 12 GB | does not fit | 11.6 GB |
| RTX 4060 Ti 16 GB | tight (encoders offloaded) | 15.4 GB |
| RTX 3090 24 GB | fits (encoders offloaded) | 23.4 GB |
| RTX 4090 24 GB | fits (encoders offloaded) | 23.4 GB |
| RTX 5090 32 GB | fits (encoders offloaded) | 31.0 GB |
| L40S 48 GB | fits | 44.0 GB |
| A100 80 GB | fits | 78.2 GB |
| H100 80 GB | fits | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | fits | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | fits | 107 GB |
| H200 141 GB | fits | 138 GB |
| B200 180 GB | fits | 176 GB |
From the model card
Loading note. This repo keeps ideogram's original FUSED attention layout (
attention.qkv/attention.o). Released diffusers expects the splitto_q/to_k/to_v/to_out.0, so a plainfrom_pretrainedhere silently random-initialises the attention weights and produces garbage images rather than raising. Use unsloth/ideogram-4-nf4-diffusers for the diffusers-native layout, or a loader that splits the fused qkv itself (Unsloth Studio does this). The text encoder config also needs transformers 5.x. Both behaviours are inherited from upstream, not introduced by this mirror.
Ungated mirror of ideogram-ai/ideogram-4-fp8, republished by Unsloth so it can be downloaded without a Hub gate. The weights are unmodified. See NOTICE and LICENSE in this repo for the terms, which are the upstream terms and are unchanged.
Ideogram 4 is Ideogram's first open weight text-to-image model. It is a state-of-the-art foundation model trained from scratch — not a fine-tune of any existing model. It introduces a new structured JSON prompting interface, with best-in-class multilingual text rendering, deep language understanding, explicit bounding-box layout and color-palette controls, and native 2k resolution images. The easiest way to try the model is online at ideogram.ai.
We believe openness drives innovation, and we invite the research community to innovate with us on the forefront of visual intelligence.
| Model | Params | Weight Quantization | Supported Hardware | Diffusers Support | License |
|---|---|---|---|---|---|
| Ideogram 4 (nf4) | 9.3B | nf4 | CUDA | Yes | Ideogram 4 Non-Commercial |
| Ideogram 4 (fp8) | 9.3B | fp8 | All | No | Ideogram 4 Non-Commercial |
We plan to support more quantizations in the future.
We evaluate Ideogram 4 across third-party arenas and benchmarks, standard open-source benchmarks, and our own internal human-preference benchmark. Across all of them, Ideogram 4 is the best open-weight image model by far, and sits at the frontier of design.
Design Arena is a third-party image Elo leaderboard focused specifically on design-oriented generation. On the overall board, Ideogram 4 is the top-ranked open-weight model, trailing only proprietary GPT and Gemini models:
Filtered to open-weight models only, Ideogram 4 leads by a commanding margin, well ahead of the next-best open model:
ContraLabs ran a blind typography evaluation judged by ten professional designers from Contra's top-earning talent. Ideogram 4 leads on first-place win rate, picked as the best of four models 47.9% of the time overall — well ahead of Gemini 3.1 Flash Image Preview (Nano Banana 2) at 30.0%, FLUX.2 [max] (15.5%), and Grok Imagine 1.0 (15.0%):
It also wins on practical usability: asked "Would you use this in real client work?", the same designers rated Ideogram 4 highest at 3.55 / 5 — significantly above Nano Banana 2 (2.84), Grok Imagine 1.0 (2.61), and FLUX.2 [max] (2.49):
On LMArena, a third-party text-to-image leaderboard that measures general-purpose text-to-image use cases, Ideogram is the top-ranked open-weight lab and a top-5 image generation lab overall — beaten only by giant companies with vastly larger budgets and resources:
For our internal human-preference benchmark, focused on graphic design and photography, we had graphic designers deeply familiar with professional design work do the rating blind. Bradley-Terry scores rank Ideogram 4 #2 overall — behind only GPT Image 2 medium — and the top open-weight model:
On standard open-source benchmarks measuring core capabilities — layout control (7Bench), spatial reasoning and object fidelity (SpatialGenEval), text rendering (X-Omni OCR), and prompt alignment (Prism) — Ideogram 4 closes the gap to the leading closed-source models across every axis. On layout control (7Bench), it is significantly better than all closed-source models:
At 9.3B parameters, Ideogram 4 delivers the best text rendering of any open-weight release we benchmarked — ahead of much larger models like Qwen-Image (20B), FLUX.2 [dev] (32B), and HunyuanImage 3.0 (80B MoE):
The inference code lives in the ideogram4 GitHub repo. Clone it, then from the repo root:
pip install .
If you plan to modify the code, install in editable mode instead so changes
under src/ideogram4/ take effect without reinstalling:
pip install -e .
The model weights are gated on Hugging Face, so you must accept the gate and
authenticate before the code can download them — otherwise the download fails
with a 404 / GatedRepoError.
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.
Running it yourself
Rent a machine by the hour. ComfyUI is installed on it. Open ComfyUI through the tunnel, load the workflow from the model's card on Hugging Face, and choose this file in its text encoder loader.
# on your rented machine: pip install diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline
pipe = DiffusionPipeline.from_pretrained("unsloth/ideogram-4-fp8", torch_dtype=torch.bfloat16).to("cuda")
image = pipe(prompt="a red bicycle on a cobbled street").images[0]
image.save("/workspace/out.png")# on your rented machine (the ssh line is on its page in the console)
# get REPO FILE FOLDER: one file into /workspace/models/FOLDER, where ComfyUI loads it from
get() { hf download "$1" "$2" --local-dir /workspace/hf-files && mkdir -p "/workspace/models/$3" && mv "/workspace/hf-files/$2" "/workspace/models/$3/$4"; }
# the model (8.2 GB)
get unsloth/ideogram-4-fp8 ideogram-4-text_encoder-FP8.safetensors text_encoders
start-comfyui# on your computer, in a second terminal: ComfyUI in your browser at http://localhost:8188
# HOST, PORT and BOOKING are your machine's, from its page in the console
ssh -o HostKeyAlias=axforge-BOOKING -o StrictHostKeyChecking=accept-new -L 8188:localhost:8188 dev@HOST -p PORT