Skip to content

Provider Qwen

h-avoyan edited this page Jun 18, 2026 · 4 revisions

Qwen

Mode: image · Models: 3

Vendor: Qwen Cloud (Alibaba DashScope) · Official API docs: Qwen API reference

Qwen (by Alibaba) is a text-to-image family with strong typography and prompt-following. The base Qwen 2 generates up to 1K with optional image input; Qwen 2 Pro adds 2K resolutions, a negative prompt, and prompt enhancement; Qwen Edit Plus is a dedicated image-edit model that composes from up to 3 source images.

Models

id Name Input type
qwen-image-2 Qwen 2 t2i
qwen-image-2-pro Qwen 2 Pro t2i
qwen-image-edit-plus Qwen Edit Plus i2i

CLI

# text-to-image
gen-ai generate -m qwen-image-2 \
  -p "a neon ramen shop sign at night, bold typography, rain reflections" -n 4

# 2K with a negative prompt
gen-ai generate -m qwen-image-2-pro \
  -p "minimalist product poster, large serif headline" \
  --neg "clutter, watermark" -r 2048x2048

# multi-image edit: compose from up to 3 sources
gen-ai generate -m qwen-image-edit-plus \
  -p "place the product on the marble table, soft daylight" \
  -i ./product.png -i ./table.jpg

MCP

{ "name": "picsart_generate",
  "arguments": {
    "model": "qwen-image-2",
    "prompt": "a neon ramen shop sign at night, bold typography, rain reflections",
    "count": 4
  } }
{ "name": "picsart_generate",
  "arguments": {
    "model": "qwen-image-2-pro",
    "prompt": "minimalist product poster, large serif headline",
    "negativePrompt": "clutter, watermark",
    "resolution": "2048x2048"
  } }

Parameters — qwen-image-2

Param CLI flag Type Values
prompt -p text required
count -n enum 1 · 2 · 4 · 6 · 8 · 10 (default 1)
imageUrls -i file optional reference image (max 1)

Source: gen-ai models info qwen-image-2 --json.

Parameters — qwen-image-2-pro

Param CLI flag Type Values
prompt -p text required (max 800 chars)
negativePrompt --neg text what to avoid
resolution -r enum 2048x2048 · 2688x1536 · 1536x2688 · 2368x1728 · 1728x2368 (default 2048x2048)
count -n enum 1 · 2 · 4 · 6 (default 1)
enhancePrompt --enhance-prompt boolean rewrite the prompt for richer detail (default true)
imageUrls -i file optional reference image(s) (max 3)

Source: gen-ai models info qwen-image-2-pro --json. qwen-image-edit-plus takes prompt (required) and imageUrls (required, up to 3 source images).

Pricing

gen-ai pricing qwen-image-2 -n 4

Cost scales with the number of images (count); on Qwen 2 Pro the chosen resolution is the other driver.

Clone this wiki locally