Alibaba · image model
0.4 credits / image · every plan · price breakdown · API playground
Made with Qwen Image 3 · Skin portrait
Qwen Image 3 is Alibaba’s latest open image model — two major versions past the original Qwen Image — combining generation and instruction-based editing with the family’s trademark reliable text rendering.
Original results
These examples use media produced for Made. Prompts are included when the production record allows them.
“Close portrait of a woman with dewy skin and freckles in soft window light, minimal makeup, calm expression. 35mm film still, soft natural light, muted earth tones with one peach accent, shallow depth of field, subtle film grain, candid and unposed, no text, logos, or watermark.”
“Amber glass serum bottle on travertine beside a shallow pool of water, label reading exactly "DEW" with a smaller line "Daily Serum". Clean editorial design, muted earth tones with one amber accent, subtle paper texture, sharp legible type, no other words or logos.”
“Bathroom shelf with skincare jars, a sprig of eucalyptus and a folded linen towel, morning light. 35mm film still, soft natural light, muted earth tones with one eucalyptus green accent, shallow depth of field, subtle film grain, candid and unposed, no text, logos, or watermark.”
“Macro of three lipstick bullets in rust, rose and plum resting on honed stone. 35mm film still, soft natural light, muted earth tones with one rust accent, shallow depth of field, subtle film grain, candid and unposed, no text, logos, or watermark.”
“Minimalist skincare box on a paper backdrop with type reading exactly "SOFT RITUAL" and a smaller line "Night Cream". Clean editorial design, muted earth tones with one lavender accent, subtle paper texture, sharp legible type, no other words or logos.”
“Stacked river stones and a folded towel with rising steam in a stone spa room. 35mm film still, soft natural light, muted earth tones with one slate blue accent, shallow depth of field, subtle film grain, candid and unposed, no text, logos, or watermark.”
These are the controls exposed by Made, which may be narrower than a provider’s full API.
Describe the subject and composition; put exact display text in quotes — text rendering is the family strength.
Upload a reference and describe the change to route to the editing endpoint.
Choose square, landscape, or portrait; output renders at the 1K tier.
Check text accuracy and adherence, then adjust one constraint at a time.
0.4 credits / image — the full breakdown, verified.
6 tested prompts with the outputs they produced.
Side-by-side prices and specs, with a straight verdict.
Side-by-side prices and specs, with a straight verdict.
Side-by-side prices and specs, with a straight verdict.
Side-by-side prices and specs, with a straight verdict.
Side-by-side prices and specs, with a straight verdict.
Side-by-side prices and specs, with a straight verdict.
image
Muse Image is Meta’s first image model, built for precise edits that change only what you ask, coherent multi-turn editing, multi-reference composition, and accurate rendered text — at 1 credit per image.
Explore model →image
Seedream 5 Lite is the fast tier of ByteDance’s Seedream 5 image family, generating and editing photoreal images at a flat rate below Nano Banana and Seedream 4.5.
Explore model →image
Nano Banana 2 runs Google’s Gemini 3.1 Flash Image, generating and editing images from 0.5K to 4K in 15 aspect ratios: fast and affordable for everyday work.
Explore model →Create scroll-stopping product photos and marketing visuals in seconds. Describe your shot, pick your style, and get professional images ready for your store, social feed, or ad campaign.
See the tool →Skip the complex menus and toolbars. Edit Images With Text Prompts. Our AI photo editor understands what you want. Type "remove the background" or "make it brighter" and watch it happen instantly.
See the tool →Upload an image, pick a size or type exact pixels, and choose whether Made crops it to fill, fits it inside, or stretches it.
See the tool →It is the third major version of Alibaba’s open Qwen Image model, adding instruction-based editing and stronger prompt adherence to the family’s reliable text rendering.