Chinese text, done right
Hanzi on menus, signs and scrolls come out as real characters, next to correct English. Our chef's board and the calligrapher's 福 both did.
Qwen Image · by Alibaba
Qwen Image is Alibaba's image model family, known above all for writing Chinese characters and English text correctly in the same picture, from menu boards to calligraphy, around natural-looking people. Describe an image below, or add a photo to edit, and it opens in Image Studio on Qwen Image 3.

Example image
Opens in Image Studio with the model you picked. You see the price in credits before anything runs.
Hanzi on menus, signs and scrolls come out as real characters, next to correct English. Our chef's board and the calligrapher's 福 both did.
Qwen Image 3 takes very long prompts, Alibaba says, and holds small text down to about 10 pixels high in a dozen languages.
Moved from a Tokyo street to a conference stage, our vlogger kept the face, the hair and even the logo on the jacket.
The first Qwen Image is open weights under Apache 2.0, and it stays on Affogato as the cheapest option in the family.
Every image below was generated with the prompt shown, unedited. Load any prompt into the generator and make your own version.
Chinese and English textA young chef in a white jacket in a small Chengdu restaurant holds up a chalkboard menu that reads "麻婆豆腐" in large Chinese characters and, below it, "Mapo Tofu · House Special" in English. Steam from the kitchen, warm tungsten light, he smiles at the camera.
CalligraphyA portrait of an elderly calligrapher with a white beard writing the character "福" in black ink on red paper with a large brush, in a sunlit studio full of hanging scrolls. Shallow depth of field, real skin texture.
Creators on cameraCandid vertical photo of a 24-year-old creator with a short wolf-cut filming a vlog on a rainy Tokyo street at night, holding a small camera at arm's length and grinning into it. Neon reflections on wet pavement, natural skin texture, shot on a phone.
Same face, new sceneKeep the exact same person from the first image: same face, haircut and smile. Now they are on stage at a tech conference, holding a microphone and presenting, a large screen glowing behind them.
Poster with a personA concert poster featuring a jazz saxophonist in a sharp burgundy suit mid-solo under a single spotlight. The headline at the top reads "BLUE HOUR SESSIONS". Below the image: "Friday 12 December · Doors 8 PM · The Velvet Room". Bold modern typography, film-grain texture.
Fashion on a personFull-body street-style photo of a woman in her fifties with silver hair in a tailored cobalt-blue trench coat, wide cream trousers and loafers, crossing a sunlit Paris boulevard mid-stride. Framed like a fashion-week street photographer's shot.
UGC with a labelUGC-style vertical selfie of a man in his twenties at a gym holding a matte black protein shaker up to the camera. The bottle reads "FORGE" and, below it, "30g PROTEIN" in bold white letters. Natural sweat, overhead gym lighting, slightly crooked phone framing.
Group photosFive friends in their twenties laughing on a rooftop at sunset, one pouring lemonade from a glass jug, string lights overhead. Candid documentary style, every face distinct and natural.
Start here
Most images, and bilingual text
The newest Qwen: long prompts, small text, edits that keep the person. Start here.
8 credits per image
Open with Qwen Image 3Option 2
The cheapest option
August 2025, open weights under Apache 2.0. Strong on Chinese and English text for its price.
4 credits per image
Open with Alibaba Qwen ImageQwen Image 2 and Qwen Image 2 Pro (February to June 2026) are also in the studio's model menu. Credits are read from the live catalog; the Standard plan ($24/month) includes 2,000 credits, about 250 Qwen Image 3 images.
Write what you want in the box at the top of this page, or start from an example. Add a photo to edit it.
Your prompt, photo, format and the Qwen version you picked carry over. Nothing runs until you press Generate.
The credit price shows under the Generate button before anything is spent.
Zoom in on any text before you publish, and fix a wrong word with an edit.
Writing Chinese and English text correctly in the same image, long detailed prompts, and edits that keep a person the same.
The original Qwen Image is open weights under Apache 2.0. Qwen Image 2 Pro and Qwen Image 3 are closed models.
Qwen Image 3 for most work and anything with text. The original Qwen Image when price matters most.
Alibaba's Qwen team, which also makes the Qwen language models. Alibaba's Wan video models are on Affogato too.
No. Every image costs credits, shown before you generate. Credits come with a monthly plan and unused credits roll over up to the plan's cap.
Yes. Paid Affogato plans include a commercial license, within our terms of service.
Facts checked November 15, 2026. Sources: Unite.ai: Qwen Image 3.0 launch · Build Fast with AI: Qwen Image 3 review · GIGAZINE: Qwen Image 2.0 · fal: Qwen Image 3 · fal: Qwen Image 3 Edit · Hugging Face: Qwen-Image