Ideogram 4.0 🧠 Model Open source
Ideogram's text-rendering image model with JSON prompting, now also as non-commercial open weights
- Maker
- Ideogram
- Latest release
- 2026-06
- Weights
- Open
- Licence
- Ideogram 4 Non-Commercial
- Apps offering it
- 11
Where you can use Ideogram 4.0
Apps in this directory that let you generate with Ideogram 4.0, from their own documentation.
Good for
About Ideogram 4.0
What it is
Ideogram is a Toronto-based image model company known for rendering text inside images. Ideogram 4.0 was released on 3 June 2026. It is a 9.3B single-stream diffusion transformer trained from scratch, and for the first time Ideogram published the weights on Hugging Face under a non-commercial licence.
Key features
- Multilingual text rendering for logos, signage, packaging and posters
- Trained on structured JSON captions: define each text element with its own position and style
- Bounding-box layout and colour-palette control (up to 16 hex colours per image, 5 per element)
- Magic Prompt turns plain descriptions into structured JSON for you
- Flexible resolutions from 256 to 2048 px, aspect ratios up to 6:1
- Edit, remix, reframe and background replacement on the API
Where you can use it
On the Ideogram web app, the Ideogram API (Ideogram 4.0 and 3.0 endpoints) and Ideogram's MCP server for assistants such as Claude Code and Cursor. Krea offers Ideogram 4.0, ComfyUI has Ideogram partner nodes (including V4), and InvokeAI can run the Ideogram 4 weights locally.
Pricing and rights
The web app has free and paid plans with priority and slow credits; paid plans add private generation, uploads and batch generation. Top-up credits roll over for a year. Ideogram's live pricing page blocked our fetch, so we list no plan price. API billing is separate from subscriptions and varies by endpoint and model version. The open weights are licensed for non-commercial use only; for commercial work, use the paid app or API and follow Ideogram's terms.
Who it is for
Designers, marketers and small businesses who need accurate text in images: logos, social graphics, ads and merch mockups.
Verdict
Ideogram 4.0 remains a specialist for typography with unusually precise layout control through JSON prompting. The open weights are for experimentation only, and it has fewer app integrations than the biggest model families.
Pros
- Multilingual text rendering for logos, signage and posters
- JSON prompting with bounding boxes and up to 16 hex colours for exact layouts
- Magic Prompt converts plain text into structured JSON
- Open 9.3B weights you can run locally for personal projects
Cons
- Open weights are non-commercial only
- Native resolution tops out at 2048 px
- Ideogram's pricing page blocked verification, so plan prices are unconfirmed
- Fewer app integrations than FLUX or Nano Banana
Similar AI models
All image models →GPT Image 2.5 🧠 ModelFreemium
OpenAI's image model behind ChatGPT Images, with 4K output and up to 16 reference images
Nano Banana 2 🧠 ModelFreemium
Google's Gemini image models (Nano Banana 2, Pro and 2 Lite) for generation and editing up to 4K
FLUX.2 🧠 ModelOpen source
Black Forest Labs' FLUX.2 family: hosted [max]/[pro]/[flex] plus open-weight [dev] and [klein]
Midjourney V8.2 🧠 ModelPaid
Midjourney's own image model, known for aesthetics and personalization, now in V8.2
Seedream 5.0 Pro 🧠 ModelFreemium
ByteDance's Seedream image models, used in CapCut, Dreamina and many creative apps
Qwen-Image 3.0 🧠 ModelOpen source
Alibaba's Qwen image models, known for dense multilingual text rendering, with open-weight versions