Skip to main content

Qwen Image 2.0: AI Image Generator & Editor Online

AI Image Generator & EditorFrom 10 credits★ 4.1 / 5Flexible content filterUpdated October 2026
Generate with Qwen Image 2.0

Bottom line

Qwen Image 2.0 generates from text and edits with up to 3 reference images in one model, up to 1920x1080, for a flat 10 credits. Generate from text or edit an existing image with Qwen Image 2.0 in one place. VdoBloom shows the credit cost for either mode before you confirm.

Example output generated with Qwen Image 2.0

Qwen Image 2.0 is an Alibaba Qwen image model that both generates and edits: on VdoBloom you can type a prompt to make a new image, or upload up to three reference images and describe the change, and both jobs cost a flat 10 credits. Output tops out at 1920x1080 (full HD) in 16:9 or 9:16, across seven aspect ratios, and there is no resolution setting to choose. It is for people who want one predictable price and one model for a generate-then-tweak workflow, and it is also the model VdoBloom's Auto Smart Edit falls back to when its first editors fail.

Is it still worth 10 credits? Newer Qwen tiers are cheaper per image: Qwen Image 3.0 is 5 credits flat at 1K or 2K and Qwen Image 2.1 is 3 credits at 1K. The case for Qwen Image 2.0 is that it is a single model for both jobs, with a 2,000-character prompt and a fixed full-HD frame for widescreen and vertical work. The sections below put those trade-offs in tables so you can pick quickly.

There is no subscription requirement. New accounts get 10 free credits, exactly one Qwen Image 2.0 job. A one-time 300-credit pack is $19.50 and those credits never expire, which is 30 Qwen Image 2.0 images; the Lite plan is $15 a month for 300 credits that reset each cycle.

Qwen Image 2.0 specs at a glance

These rows come from VdoBloom's image configuration and the Runware service that sends each job. The output size is fixed per aspect ratio, not chosen by a resolution setting, and the largest frames are the 16:9 and 9:16 ones.

SpecQwen Image 2.0 on VdoBloom
TypeText-to-image and image editing in one model
Aspect ratios1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3
Output size per ratio1920x1080 (16:9), 1080x1920 (9:16), 1280x1280 (1:1), 1280x960 (4:3), 960x1280 (3:4), 1200x800 (3:2), 800x1200 (2:3)
Resolution settingNone; size follows the ratio
Reference images (edit)1 to 3; uploads are resized to 240-2048 px per side before sending; 10 MB per file
Prompt lengthThe primary route reads up to 2,000 characters
Price10 credits flat per job, generate or edit
ProvidersRunware first, then Together AI, then a KIE Qwen2 fallback
Content labelHOT, VdoBloom's more permissive filtering tier; platform rules still apply

What Qwen Image 2.0 costs: worked examples

One number covers everything: 10 credits per job. Generating from text, editing with one reference image or editing with three all cost the same, and so does every aspect ratio. The dollar column uses the Lite plan rate of $0.05 a credit ($15 for 300). The cost is shown before you generate and a failed job refunds automatically.

JobsCreditsAt $0.05 a credit
1 generation or edit10 cr$0.50
Generate 1 + 2 edit passes30 cr$1.50
5 jobs50 cr$2.50
10 jobs100 cr$5.00
Whole 300-credit pack30 jobs$15.00 on Lite, $19.50 one-time
10 free signup credits1 job, 0 leftfree

Read more: Credit packs and plans · How to use Qwen Image 2.0, step by step

Qwen Image 2.0 vs Qwen Image 2.0 Pro vs Qwen Image 3.0 vs Qwen Image 2.1

These are the Qwen choices people weigh against each other. Qwen Image 2.0 and its Pro tier are the two that generate and edit in one model; the 3.0 and 2.1 lines split those jobs into separate generator and editor models with a 1K or 2K setting. Per image, the newer lines cost less.

ModelPriceMax sizeRatiosEdits in same modelPrompt cap
Qwen Image 2.010 cr flat1920x10807Yes, up to 3 references2,000
Qwen Image 2.0 Pro12 cr flat1920x10807Yes, up to 3 references2,000
Qwen Image 3.05 cr at 1K or 2K2048 px long side8 (incl. 21:9)No; Qwen Image 3.0 Edit is separate (5 cr)800
Qwen Image 2.13 cr at 1K, 5 at 2K2K9 (incl. 21:9, 9:21)No; Qwen Image 2.1 Edit is separate (3/5 cr)5,000

Read more: Qwen Image 2.0 Pro · Qwen Image 3.0 · Qwen Image 2.1 · Qwen Image 3.0 Edit

Standard or Pro: what the extra 2 credits buy

Qwen Image 2.0 and Qwen Image 2.0 Pro run through the same service with the same seven ratios, the same fixed sizes, the same three-reference edit limit and the same 2,000-character prompt. The Pro tier is described in VdoBloom's picker as enhanced visual fidelity and typography, and the routing guide reserves it for typography-heavy designs such as flyers and banners, while the standard tier is listed for fast, casual images and concepts.

If the image has no words in it, the standard tier at 10 credits is the sensible default. If a headline or label has to read cleanly, the 2-credit step to Pro is the cheaper fix compared with re-running the standard model several times.

Read more: How to use Qwen Image 2.0 Pro

Where to run Qwen Image 2.0 on VdoBloom

Qwen Image 2.0 appears in two places in the dashboard, and the price is 10 credits in both.

  • Image Generator (Dashboard, Images, Generate): pick Qwen Image 2.0, write the prompt, choose one of the seven ratios and generate. There is no resolution control; the ratio sets the size.
  • Image Edit tab: pick Qwen Image 2.0, upload one to three reference images, describe the change and choose the output ratio. Uploads are resized into the 240-2048 pixel range automatically, so oversized phone photos are accepted.
  • Auto Smart Edit: when the Auto edit model's first choices (Seedream 5.0 Pro, Seedream 4.5, WAN 2.7) all fail, the chain moves on to WAN 2.7 Pro and then Qwen Image 2.0. Auto bills its own 7-credit rate, not Qwen Image 2.0's 10.
  • Not here: transparent backgrounds and painted-mask edits. For those, use Qwen Image 2.1 Edit.

Read more: AI image editor from a photo: every edit model compared · Auto Smart Edit

Getting good results from one model that does two jobs

Because the same prompt box handles both modes, write for the mode you are in. To generate, describe the whole scene: subject, setting, light, framing. To edit, describe only the change ("replace the sky with a sunset, keep the people as they are") and let the reference images carry everything else. A full scene description on an edit job tends to repaint the parts you wanted to keep.

With several references, say which image does what: "use the jacket from the second image on the person in the first image". Pick the ratio for the placement before you generate, since there is no resolution setting to fall back on; 16:9 and 9:16 give the largest frames at 1920x1080, while 3:2 and 2:3 come out smaller at 1200x800. If a picture needs to be larger than full HD, Qwen Image 3.0 at 2K costs half as much per image.

  • Generate: "A neon-lit ramen stall at night in the rain, steam rising, 35mm street photo, 16:9."
  • Edit: "Change the stall sign to read 'OPEN LATE' in red neon; keep everything else unchanged."
  • Multi-reference: "Put the cat from image 2 on the counter in image 1, matching its lighting."

Which Qwen model to pick for your job

Qwen Image 2.0 is the right pick for some jobs and the wrong one for others. This list matches common jobs to the Qwen model that fits them best on price and features, using only the specs and credit costs on this page.

  • One model for generate and edit, same price for both steps: Qwen Image 2.0 (10 credits) or, if the image has a headline, Qwen Image 2.0 Pro (12).
  • Lots of text-free drafts on a small budget: Qwen Image 2.1 at 1K (3 credits) or Qwen Image 3.0 (5 credits at 1K or 2K).
  • An image larger than full HD, or a square bigger than 1280 px: Qwen Image 3.0 at 2K (2048x2048 square, 5 credits).
  • A product cut-out on a transparent background: Qwen Image 2.1 or Qwen Image 2.1 Edit with transparency on (3 credits at 1K).
  • Editing with more than three reference images: Qwen Image 3.0 Edit or Qwen Image 2.1 Edit, both up to ten.
  • Changing one painted area and leaving the rest untouched: Qwen Image 2.1 Edit in mask mode.
  • Realism plus typography at 2K: Qwen Image 3.0 Pro (7 credits at 2K).

Read more: Qwen Image 2.1 Edit (mask and transparency) · Qwen Image 3.0 Pro

Limits and honest caveats

The ceiling is 1920x1080 and only in 16:9 or 9:16; square images are 1280x1280 and 3:2 frames are 1200x800, so for print-size work the 2K tiers of Qwen Image 3.0 or 2.1 give more pixels for fewer credits. Edits take three reference images at most, where Qwen Image 3.0 Edit and Qwen Image 2.1 Edit take ten. There is no negative-prompt field, no transparent background and no mask mode. The primary route reads the first 2,000 characters of a prompt. If Runware and Together AI both fail, the last fallback sends the job to KIE's Qwen2 model rather than Qwen Image 2.0 itself, which keeps the job from failing but can change the look. Like every model on VdoBloom it is hosted, not downloadable, and a job that fails on every route refunds its credits.

VdoBloom does not publish quality scores for models. The value question here is about workflow, not looks: one model for generate and edit at 10 credits, against cheaper split models at 3 to 5.

What creators use Qwen Image 2.0 for

  • Generating a scene and then refining it with up to three reference images in the same model, at the same flat price for both steps.
  • Full-HD 1920x1080 thumbnails and headers, or 1080x1920 vertical story frames, at native size with no resolution choice to make.
  • Combining a product photo with a style or background reference (up to three images) in a single instruction edit.
  • Budgeting by count: every job is 10 credits, so 30 jobs are 300 credits whatever the ratio or edit.

How to use Qwen Image 2.0 on VdoBloom

  1. 1Open the Text to Image tool and select Qwen Image 2.0.
  2. 2Write your prompt (or upload an image to edit).
  3. 3Set aspect ratio, review the credit cost, and generate.

Frequently asked questions

How much does Qwen Image 2.0 cost?

A flat 10 credits per job, whether you generate from text or edit with one to three reference images, and whatever the aspect ratio. At the Lite rate of $0.05 a credit that is $0.50 a job; 300 credits buy 30 jobs.

Can I try Qwen Image 2.0 for free?

Yes. New accounts get 10 free credits, which is exactly one Qwen Image 2.0 generation or edit. If you want several tries on the free balance, Qwen Image 2.1 costs 3 credits at 1K, so the same 10 credits cover three images.

Can Qwen Image 2.0 edit images, or only generate them?

Both. In the Image Generator it creates images from a prompt; in the Image Edit tab it takes one to three reference images plus an instruction. Both modes cost 10 credits.

What resolution does Qwen Image 2.0 output?

The size is fixed by the aspect ratio: 1920x1080 for 16:9, 1080x1920 for 9:16, 1280x1280 for 1:1, 1280x960 for 4:3, 960x1280 for 3:4, 1200x800 for 3:2 and 800x1200 for 2:3. There is no 1K/2K/4K setting.

Which aspect ratios does Qwen Image 2.0 support?

Seven: 1:1, 4:3, 3:4, 16:9, 9:16, 3:2 and 2:3. All seven cost the same 10 credits.

How many reference images can I use with Qwen Image 2.0?

Up to three per edit. Each upload can be up to 10 MB and is resized into the 240-2048 pixel range before it is sent. For more references, Qwen Image 3.0 Edit and Qwen Image 2.1 Edit take up to ten.

Qwen Image 2.0 vs Qwen Image 3.0: which is better value?

Per image, Qwen Image 3.0 is half the price (5 credits at 1K or 2K against a flat 10) and reaches 2048 pixels on the long side. Qwen Image 2.0's advantage is that one model generates and edits, with a 2,000-character prompt against 3.0's 800; Qwen Image 3.0's edits run in the separate Qwen Image 3.0 Edit model.

Qwen Image 2.0 vs Qwen Image 2.0 Pro: what is the difference?

Same sizes, ratios, prompt length and three-reference edit limit. Pro costs 12 credits against 10 and is the tier for enhanced fidelity and typography, so it suits flyers, banners and anything with a headline; the standard tier suits text-free images.

Is Qwen Image 2.0 censored?

It carries VdoBloom's HOT label, the more permissive filtering tier used for all Qwen models, against STRICT on GPT Image and Nano Banana. Platform rules still apply, and it is not a promise of explicit output.

Is Qwen Image 2.0 open source?

Alibaba has released open-weight Qwen image checkpoints, but on VdoBloom Qwen Image 2.0 runs as a hosted model through Runware (with Together AI and KIE as fallbacks). There is nothing to download and no provider key to manage.

Qwen Image 2.0 guides, prices and comparisons

This page is the Qwen Image 2.0 hub. Each guide below goes deeper on one question and links back here.

Related models