All models

Create an image

Qwen Image 3

Qwen Image 3 — flagship open image model, strongest prompt adherence and readable in-image text, up to 2K.

The lettering specialist — it renders readable words inside the image where most generators produce near-text. The pick for posters, thumbnails, and mockups that need real typography.

  • Embeds legible text in any language at scale
  • Maintains text fidelity in non-square aspect ratios (e.g., 9:16, 21:9)
  • Outputs native 1K and 2K resolution images without upscaling artifacts

from 5 credits per result · ~12s per run

1K2K1:11:22:11:3
Open Qwen Image 3 in the studio
Qwen Image 3 sample output, generated inside LivePair

How it works

  1. 1Type what you want — subject, scene, light.
  2. 2Pick a size and aspect ratio.
  3. 3The result lands in your browser's local library — download it or it's gone in 48 hours.

Prompts it takes

  • “a lighthouse at golden hour, waves crashing, cinematic”
  • “portrait of a boxer mid-training, dramatic rim light”

Load one into the studio — create an image is Qwen Image 3's job.

Call it from an agent

x402 or a prepaid key — no account needed for agent calls.

curl -X POST https://livepairai.com/v1/agent/generate \
  -H "content-type: application/json" \
  -d '{"model":"qwen-image-3","prompt":"..."}'

FAQ

What is Qwen Image 3?
Qwen Image 3 — flagship open image model, strongest prompt adherence and readable in-image text, up to 2K.
How much does Qwen Image 3 cost?
From 5 credits per result — per-use billing, no subscription. 100 credits = $1.
Is Qwen Image 3 private?
Results land in your browser's local library and auto-delete after 48 hours; nothing is published unless you share it.