Posters, labels and banners with legible words
No company's docs show a full fit yet. The listings below do the job with a limit, or not at all.
What counts as done
The given copy rendered as text in the image, including non-Latin scripts the user asked for.
Measured, for our own tests: every word is spelled correctly and readable at the output size, in at most two attempts. No docs we read state these figures, so they don't count against a listing here.
How agents ask for it: make a poster that says
, put this headline on the image
, make a label in Chinese
, banner with our slogan
Does the job, with a limit 14
It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.
| Listing | Why |
|---|---|
| Black Forest Labs FLUX API B Black Forest Labs Inc. | No supported-scripts statement, and acc-25 reports every FLUX model renders Chinese badly. source |
| Bria API BB Bria Artificial Intelligence Ltd. | /rewrite_text changes text inside an image, but no page states which scripts render correctly. source |
| fal image models BB Features & Labels, Inc. | Typography models (Ideogram, Recraft) are hosted, and acc-10 routes text jobs to them, but script support is whatever each vendor states; fal adds nothing. source |
| FLORA B FLORA AND FAUNA FLOWERSHOP, INC. | Ideogram 3.0, 4.0 and 4.5 are hosted for typography, but no script support is stated. source |
| Ideogram API D Ideogram AI, Inc. | Ideogram markets crisp typography and acc-10 and acc-14 route text jobs to it, but no page lists supported scripts or non-Latin behaviour. source |
| Iris-3B C Speridlabs | The card's limitations say text rendering is not always reliable, the Turbo checkpoint is weaker on dense text, and benchmarks are English only. source |
| Leonardo.Ai API F Leonardo Interactive Pty Ltd | Ideogram 3.0, 4.0 and 4.5 are callable for typography, but no script support is stated on Leonardo's pages. source |
| OpenAI Image API BB OpenAI OpCo, LLC | Higher complexity helps dense text, but the guide's limitations say the model can still struggle with precise text placement and clarity, and no supported scripts are listed. source |
| Qwen-Image 2.1 (Model Studio) B Alibaba Cloud (Singapore) Private Limited | The card claims improved typography, but acc-14 found the text in Qwen-Image 2.1's own samples garbled. source |
| Recraft API D Recraft Inc. | V3's text_layout places exact words, but allows uppercase Latin letters, digits and basic punctuation only; no non-Latin script support is stated for any model. source |
| Replicate image models B Replicate, LLC | Typography models such as Recraft and Ideogram are hosted, but script support is whatever each model's vendor states. source |
| Seedream on BytePlus ModelArk C BytePlus Pte. Ltd. | Seedream 5.5 can return OCR of text layers with font matching, but no page states which scripts generation renders correctly. source |
| Stability AI Image API F Stability AI Ltd | No typography or script statement in the API reference; text rendering is left to the model with no stated support for non-Latin scripts. source |
| SurgePix API F SurgePix | Poster, deck and Xiaohongshu skills render copy and the examples include Chinese, and a translate skill rewrites text in an image, but no accuracy or script statement is published. source |
Couldn't check 1
The docs were closed to us, or silent. These carry no mark either way.
| Listing | Why |
|---|---|
| Adobe Firefly API D Adobe Inc. | The API reference says nothing about text rendering or scripts, and no account names Firefly. |
What stops agents finishing
- model garbles non-Latin text
- poster with text fails in vector mode
Around it
Before: llm.write-copy, translation.translate. After: image-generation.product-and-ad-creative.
Where this comes from
Weight 8 of 100 in Image generation: 4 of 37 accounts; garbled Qwen text against Ideogram, poor Chinese on FLUX, SVG posters failing, a router sending typography to Ideogram
The same as data: /api/v1/use-cases/image-generation.text-in-image.json, or the MCP tool get_use_case with id image-generation.text-in-image. Other jobs in this category: /api/v1/niches?category=image-generation.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.