ZMIME
Image model comparison · 2 models · Updated Nov 2, 2026

Qwen-Image 2.1 vs GPT Image 2.5 Flare

Preferred for text to image: GPT Image 2.5 Flare · Preferred for image editing: GPT Image 2.5 Flare · Newest: Qwen-Image 2.1. Quality scores are blind human-preference votes; specs and prices come from each lab’s listings.

  1. Alibaba (Qwen)

    Qwen-Image 2.1

    Released Sep 14, 2026New

    • Text to image
    • Image editing
    • Arena1,362
    • From—
    • Max size—
    Full specs Remove Qwen-Image 2.1

  2. OpenAI

    GPT Image 2.5 Flare

    Released Sep 8, 2026New

    • Text to image
    • Image editing
    • Arena1,482
    • From—
    • Max size—
    Full specs Remove GPT Image 2.5 Flare

  3. Add a model

    Make it a three-way comparison.

01 — At a glance

Where they differ

Quality is the arena score from blind, side-by-side human votes. Price, length and resolution come from each lab’s own listings.

  • Preferred for text to imageGPT Image 2.5 FlareQwen-Image 2.1 1,225 (#19) · GPT Image 2.5 Flare 1,401 (#2) — blind human votes
  • Preferred for image editingGPT Image 2.5 FlareQwen-Image 2.1 1,362 (#20) · GPT Image 2.5 Flare 1,482 (#2) — blind human votes
  • NewestQwen-Image 2.1Qwen-Image 2.1 Sep 14, 2026 · GPT Image 2.5 Flare Sep 8, 2026
  • Widest inputsSame inputsQwen-Image 2.1: Text prompts, Reference images · GPT Image 2.5 Flare: Text prompts, Reference images
  • Image editingBoth edit imagesTakes a reference image as input: Qwen-Image 2.1 yes · GPT Image 2.5 Flare yes
  • Self-hostingQwen-Image 2.1Downloadable weights (qwen-research)
  • Most API hostsGPT Image 2.5 FlareQwen-Image 2.1: 1 · GPT Image 2.5 Flare: 4
02 — Side by side

Every spec in one table

Dashes mean the data is not published.

Qwen-Image 2.1 vs GPT Image 2.5 Flare specifications side by side
SpecificationQwen-Image 2.1Alibaba (Qwen)GPT Image 2.5 FlareOpenAI
Quality (blind human votes)
Text to image arenaElo-style score · higher is better1,225 ±6#19 of 82 · 19,604 votes1,401 ±5#2 of 82 · 39,218 votes (best)
Image editing arenaElo-style score · higher is better1,362 ±6#20 of 61 · 39,412 votes1,482 ±5#2 of 61 · 87,941 votes (best)
Price
How it is billed—Billed per token: $30 per million output image tokens.
Output
ProducesImagesImages
Images per request—Up to 10
Inputs
Text promptsYesYes
Reference images (editing)YesYes
Max reference images—16
Access
WeightsOpenqwen-researchProprietary
API hosts14 (best)
ReleasedSep 14, 2026Sep 8, 2026
03 — Questions

Which should you choose?

Which is better, Qwen-Image 2.1 or GPT Image 2.5 Flare?

For text to image, voters prefer GPT Image 2.5 Flare (Qwen-Image 2.1 1,225 (#19) · GPT Image 2.5 Flare 1,401 (#2)). For image editing, voters prefer GPT Image 2.5 Flare (Qwen-Image 2.1 1,362 (#20) · GPT Image 2.5 Flare 1,482 (#2)). Arena scores come from blind, side-by-side human votes on the same prompts; when the gap is inside both models’ error margins we call it a tie. Test your own prompts before you commit.

What is the difference between Qwen-Image 2.1 and GPT Image 2.5 Flare?

Qwen-Image 2.1 is an image model from Alibaba (Qwen) (Sep 14, 2026), taking text prompts and reference images and producing images. GPT Image 2.5 Flare is an image model from OpenAI (Sep 8, 2026), taking text prompts and reference images and producing images.

Which is newer?

Qwen-Image 2.1 Sep 14, 2026; GPT Image 2.5 Flare Sep 8, 2026.

Which can edit an existing image?

Takes a reference image as input: Qwen-Image 2.1 yes; GPT Image 2.5 Flare yes.

Can I run any of these on my own hardware?

Qwen-Image 2.1. Downloadable weights (qwen-research).

04 — Keep comparing

Related comparisons