Qwen Image 2.1: AI Image Generator Online
Bottom line
Qwen Image 2.1 adds real transparent PNG output — 3 credits at 1K, 5 at 2K, across nine aspect ratios. Pick Qwen Image 2.1 in the Text to Image tool, write your prompt, and pay per image in credits — the price is visible before you generate.
Qwen Image 2.1 is Alibaba's newest text-to-image entry on VdoBloom, added September 22, 2026, and its standout feature is real alpha-channel transparency: set the background to transparent and export as PNG or WebP, and the result is an actual cut-out you can drop onto any background — not a flat color you'd need to key out afterward.
Pricing follows resolution: 3 credits at 1K, 5 at 2K. Nine aspect ratios are supported, including the ultrawide 21:9 and tall 9:21 pairing that most Qwen tiers skip, and prompts can run up to 5,000 characters — enough for detailed scene descriptions. An enhance-prompt toggle is on by default, which rewrites short prompts into a fuller scene before generation; turn it off if you want literal control over exactly what you typed.
The model carries VdoBloom's HOT content label — more permissive filtering than the STRICT tier on GPT Image and Nano Banana — and rates 4.6 with users, ahead of the older Qwen2 (4.1) and Qwen Image 2.0 (4.1) tiers. As always, the credit cost for your chosen resolution is shown before you generate.
What it costs on VdoBloom
Qwen Image 2.1 prices by resolution: 3 credits at 1K, 5 at 2K. A session of six images — three at each resolution — costs 24 credits total. New accounts start with 10 free credits, which covers three 1K generations outright or two at 2K with credits to spare. The cost for your selected resolution is quoted before you generate, and a failed job is refunded automatically.
Qwen Image 2.1 vs the rest of the Qwen lineup
Qwen Image 2.1 sits at the cheap end of VdoBloom's Qwen ladder, tied for the top user rating (4.6) with Qwen Image 3.0. At 1K it undercuts Qwen Image 3.0's flat 5-credit price (3 versus 5); at 2K the two cost the same. Both outrate and outprice Qwen Image 2.0 (flat 10 credits, 4.1 rating) and Qwen2 Text-to-Image (8 credits, 4.1 rating), the two older tiers in the lineup.
The deciding factor between 2.1 and 3.0 is usually feature-driven rather than price-driven: 2.1 is the only one with real transparent-background export and the widest aspect-ratio set (nine, including 21:9/9:21); 3.0 has a simpler flat price across both resolutions.
Transparent backgrounds and long prompts
Transparent output requires two settings together: background set to transparent, and output format set to PNG or WebP (JPEG has no alpha channel and the combination is rejected). Describe only the subject in the prompt — no background — and the model renders a genuine cut-out rather than a solid color to remove afterward.
The 5,000-character prompt limit is generous enough for full scene descriptions with lighting, camera angle, and material detail spelled out explicitly. The enhance-prompt toggle, on by default, will fill in gaps in a short prompt automatically; turn it off in the composer if you'd rather the model follow exactly what you wrote, with nothing added.
What creators use Qwen Image 2.1 for
- Producing product cut-outs or logo assets with a genuine transparent PNG background, ready to composite elsewhere.
- Generating ultrawide 21:9 banners or tall 9:21 story graphics that most other Qwen tiers don't offer.
- Writing long, detailed prompts (up to 5,000 characters) for complex scenes, with enhance-prompt available to fill in gaps.
- Budgeting cheaply at 1K (3 credits) before committing to a 2K final render (5 credits).
How to use Qwen Image 2.1 on VdoBloom
- 1Open the Text to Image tool and select Qwen Image 2.1.
- 2Write your prompt.
- 3Set resolution (1K/2K), aspect ratio, review the credit cost, and generate.
Frequently asked questions
How much does Qwen Image 2.1 cost on VdoBloom?
3 credits at 1K resolution, 5 credits at 2K. Generating four images at 1K costs 12 credits; the same four at 2K cost 20. The resolution you pick sets the price — aspect ratio doesn't change it.
Can I try Qwen Image 2.1 for free?
Yes — new accounts get 10 free credits at signup, enough for three 1K generations or two 2K generations, with credits left over either way.
Is Qwen Image 2.1 censored?
It carries VdoBloom's HOT content label, the platform's more permissive filtering tier — noticeably more flexible than the STRICT label on GPT Image and Nano Banana. VdoBloom doesn't promise explicit or NSFW output on any model.
Does Qwen Image 2.1 support transparent backgrounds?
Yes — set background to transparent and choose PNG or WebP as the output format (JPEG has no alpha channel, so that combination is blocked). It's one of the only text-to-image models on VdoBloom with real alpha-channel transparency rather than a flat color you'd need to key out afterward.
Qwen Image 2.1 vs Qwen Image 3.0 — which should I use?
The two are both rated 4.6 by VdoBloom users. At 1K, Qwen Image 2.1 is cheaper (3 credits versus Qwen Image 3.0's flat 5); at 2K they cost the same (5 credits each). The practical difference is features: 2.1 supports real transparent-background export and nine aspect ratios (including 21:9/9:21); 3.0 covers eight ratios with no transparency option, but its flat price is simpler to budget.
Qwen Image 2.1 vs Qwen Image 2.0 — what changed?
Qwen Image 2.1 is cheaper (3-5 credits by resolution, versus 2.0's flat 10), rated higher (4.6 versus 4.1), and adds transparent-background export plus two extra aspect ratios (nine versus seven). Qwen Image 2.0's advantage is that it bundles editing into the same model with up to three reference images; Qwen Image 2.1's editing lives in the separate Qwen Image 2.1 Edit model instead.
How do I generate a transparent PNG with Qwen Image 2.1?
In the image generator, select Qwen Image 2.1, set background to transparent, and choose PNG or WebP as the output format. Describe the subject only — the model renders it without a background to key out, so there's no separate cutout step afterward.
Which aspect ratios does Qwen Image 2.1 support?
Nine: 1:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16, 21:9 and 9:21. The 21:9/9:21 pair is exclusive to this tier among VdoBloom's text-to-image Qwen models.
What's the prompt length limit for Qwen Image 2.1?
5,000 characters — the longest limit among VdoBloom's Qwen text-to-image models. The enhance-prompt toggle, on by default, will also expand a short prompt into a fuller scene description before generation if you leave it on.
Is Qwen Image 2.1 open source?
Earlier Qwen-Image checkpoints from Alibaba have been released as open weights, but VdoBloom runs Qwen Image 2.1 as a hosted API through KIE — there's no local model to download through this integration.