Nanomelon is a small independent product that turns written descriptions—and, when you want, reference photos—into still images. We built it because most “all-in-one” generators either hide every setting behind a paywall or dump you into a blank prompt box with no explanation of what the model will actually do. This page is the honest version of who we are, what the site does, and what it is not.
On the homepage you choose a mode. Text to image starts from a prompt only. Image to image starts from a picture you upload and a prompt that describes how the result should change. You pick a model, an aspect ratio, and how many variations to run. Credits are deducted only when a job completes successfully; failed jobs do not consume the balance. That rule exists because generation APIs fail for boring reasons (timeouts, policy filters, upstream capacity) and charging for those failures trains people to distrust the meter.
Outputs can be downloaded without a watermark. High-resolution originals may require a paid plan; the on-page pricing modal states the current packages. If you mark a work as publicly visible, it can appear in the public gallery so other visitors can study the prompt and try a similar look. Private works stay in your account.
A generator with no examples is a toy. A gallery of unlabeled thumbnails is not much better. We publish selected public images with the actual prompt, model name, size, and generation type so you can reverse-engineer a look instead of guessing. That is also why we write long-form guides: how to write prompts, when to use image-to-image, and how models and credits differ. The user guide covers the buttons. The learn pages cover the craft.
We are not a stock-photo marketplace and we are not a social network. Comments and open uploads are moderated. Public gallery items are staff- or user-submitted works that passed a basic policy check. If something slips through that should not be public, email us and we will take it down.
Nanomelon is operated as an independent web product at the domain nanomelon.top. Support and privacy requests go to support@flyingmob.com. We are not affiliated with the model vendors whose names appear in the model list; those names describe the upstream engine we call, not a partnership endorsement. If a vendor changes an API, a model may be renamed, paused, or replaced. We will keep the on-site labels in sync with what the generator actually calls.
Related tools we link from the footer (deep-imager, ToolPilot AI, Backgroundcanvas) are separate products. Linking them does not mean they share your Nanomelon account or your prompts.
The models we call already refuse a range of prompts: sexual content involving minors, some celebrity likenesses, brand-protected characters, and other categories in their own safety policies. On top of that, we do not want Nanomelon used for scams, non-consensual intimate imagery, or content that exists only to harass. If a job is blocked, the interface says so; credits are not taken for a blocked or failed run. Trying to jailbreak a filter by rephrasing the same request is a good way to get the account reviewed.
We also do not pretend AI output is a photograph of a real event. If you use Nanomelon for journalism, political advertising, or anything that could mislead a reader about a real person, you are responsible for labeling and for the law in your country. Our Terms of Service say the same thing in legal language.
If this site shows ads, they will sit on pages that already have real content: guides, about, this studio homepage, and similar. We will not put ads on empty shells, pop-up-only views, or pages whose only job is to host a unit. Advertising partners (including Google, if AdSense is approved) may use cookies as described in our Privacy Policy. We will not tell you to click ads, and we will not click them ourselves.
Start with a subject, a setting, a lighting note, and an aspect ratio that matches how you will use the file. “A girl, sunshine” is a mood, not a brief. “A young woman with blonde hair, backlit by late-afternoon sun, shallow depth of field, 3:4 portrait” is a brief. The prompt guide walks through that difference with examples from our own gallery. If you already have a photo of a product or a character, use image-to-image and describe only the change you want—background, wardrobe, time of day—rather than restating the entire scene.
Credits are not unlimited free generation. New accounts receive a small starter balance (currently five credits) so you can test the pipeline. Each successful Generate uses the amount shown on the button. Failed and policy-blocked jobs do not deduct credits. When the starter balance is gone, buy more through pricing or stop. Browsing Learn guides and Public Images does not cost credits. High-resolution packages may require a paid plan as stated in pricing.
Questions about the product, takedowns, or this page: Contact · support@flyingmob.com