Generate Image Mode: How can the mode where you use referenced image has a character limit??? Most AI aggregator websites don't have this limit. Why does yours have it???

Hi ​@Eugene_Fitz,

Thanks for your feedback.

The character limit is a known constraint and your frustration is understood.

Firefly's standard Text to Image prompt limit is approximately 1,000 characters. In Firefly Boards the limit varies by model — GPT Image 1.5 allows up to around 5,000 characters, which may be worth exploring for more detailed prompts.

It is worth noting that there is a technical reason for prompt limits beyond simply an arbitrary design decision. AI image generation models process prompts via a text encoder that has a fixed context window — a maximum number of tokens it can process at once. Prompts that exceed this window are either truncated or cause degradation in how accurately the model follows the later parts of the prompt. So a very long prompt does not necessarily produce better results than a well-constructed shorter one — the model may effectively ignore content beyond a certain point regardless of the character limit in the UI.

Adobe has received many requests to increase the limit and there is an active feature request thread you can upvote here: https://community.adobe.com/feature-requests-405/p-character-limit-restricts-image-descriptions-merged-thread-1479163.

Your feedback has been noted.

    droopy