Seedream 5.0 Flash builds images from a description or from up to ten of your own references, in 1K, 1.5K or 2K, and the credit quote does not change with the tier you pick.
One selector entry, two starting points. Text to image builds a picture from a description; image to image starts from one to ten uploaded references. Both share the same tier, shape and file-type controls, and both need a description.
Choose 1K, 1.5K or 2K for the size your layout needs. All three currently have the same per-image quote.
Choose a size tier, a shape and PNG or JPEG output. In editing mode, add one to ten references and the same controls apply. A description is required in both modes and must contain between 4 and 5000 characters.
Select 1K, 1.5K or 2K, then choose an aspect ratio. Review the generated file for your intended placement.
Upload one to ten JPEG, PNG or WebP references, each no larger than 30 MB. Resize or compress larger files before uploading.
1:1, 4:3, 3:4, 16:9, 9:16, 2:3, 3:2 and 21:9, selectable in text mode and in editing mode alike.
Use the same description with different size and shape settings for a header, card or poster. Each generation is a separate result.
Upload references for subjects, colours or a setting, then describe how you want them used.
Choose 21:9 for a wide header, 2:3 for a portrait layout or 3:2 for a landscape layout.
Use a button on this page. It opens the right workspace with this model already selected, in the language you are reading now.
Write what should happen in the shot or image. Concrete subjects, framing and action give the model more to work with than short labels.
Choose length or quality tier, shape and sound where offered, then read the credit quote for those exact settings before you generate.
Generation uses account credits. Cost depends on the model, quality and output size or duration. Reference images, video or audio may add charges in some modes. Check the workspace quote for your selected inputs and settings before generating; the pricing page lists credit packages.
See credit packagesThey are separate selectable products in the same family. Flash is the one with the 1K, 1.5K and 2K tiers and up to ten references per edit. We do not claim quality differences between family products; the selector lists them all, and the quote is what you can compare directly.
Not on this model. The workspace quotes the same per-image amount for 1K, 1.5K and 2K, and for PNG as well as JPEG. The amount is shown before you confirm, it multiplies by the number of images you request (up to four), and each image is submitted as its own task.
No separate reference charge is listed for this model here: the quote covers the task however many references between one and ten you upload. Read the quote for your exact settings before you confirm.
Between one and ten, in JPEG, PNG or WebP and up to 30 MB per file. The same tier, shape and format controls apply as in text generation.
A description shorter than four characters or longer than 5000 does not pass the check. Describe the subject, the setting and the framing instead of entering a single word.
No on both counts. There is no background control on this model and output is limited to PNG and JPEG, so pick a model with an explicit background setting or with WebP output for those jobs.
Other models on this platform that do a similar job, or that pair well with this one.
Nano Banana 2.1 covers both workflows on one model: images from a description, or edits driven by up to ten of your own reference images, at 1K, 2K or 4K.
Read the guideThe FLUX 3 family product available on this platform: text generation and reference editing at 1K, 2K or 4K, with PNG, JPEG and WebP output.
Read the guideOne of the two selectable GPT Image 2.5 products on this platform: text and reference-based image generation with transparent-background support at the 1K tier.
Read the guide