Alibaba Qwen-Image 2.0 Pro (Official Edition)

If the regular Qwen-Image 2.0 node is the workhorse, this is the one you pull out when you need the output to actually be good - Qwen-Image 2.0 Pro, the tier Alibaba sells for harder prompts, better text rendering, and higher-fidelity results. It's the same 7B API-only model family that launched without open weights in early 2026, same cloud-only situation, and this Official Edition node is your ComfyUI doorway to the vendor's official qwen-image-2.0-pro-official/text-to-image endpoint.

What's different about Pro

The node's parameter surface is nearly identical to the non-Pro edition - same prompt cap of 1300 tokens, same negative prompt, same size/watermark/prompt_extend controls. What you're paying for isn't new knobs, it's the model behind the endpoint: Pro is the higher-quality tier for when the standard pass comes back with weak hands, mushy text, or a composition that doesn't hold up. In a batch pipeline, the pragmatic play is standard for drafts and Pro for finals - though both bill against your BizyAir credits, and Pro bills more.

How it works

Pure cloud API call, same as every BizyTRD node: prompt up to api.bizyair.ai, remote generation on the vendor endpoint, result downloaded back as a native IMAGE tensor. No GPU, no model files, metered billing. The only real constraint is the 1300-token prompt ceiling, which is more room than you'll usually need.

Inputs that matter

• prompt - up to 1300 tokens. This is where Pro earns its keep: it handles long, detailed, multi-clause briefs without falling apart.
• negative_prompt - up to 500 characters of exclusions.
• size - defaults to 2048*2048.
• n - number of images; the slider is capped at 1 in this build despite the tooltip's "1–6".
• seed - integer, default 0, for reproducible runs.
• watermark - default off; flip on for a stamped Qwen-Image corner mark.
• prompt_extend - default on; turn off when your prompt must survive verbatim.
• api_config / skip_error - per-request key/base override and placeholder-on-error.

Outputs: image, request_id, urls.

Install and first run

ComfyUI Manager → "BizyAirPlus", or:

cd /path/to/ComfyUI/custom_nodes
git clone https://github.com/siliconflow/BizyAirPlus.git
cd BizyAirPlus
python -m pip install -r requirements.txt

Restart, register at bizyair.ai, set the key in Settings > BizyAirPlus > API Key (or BIZYAIR_API_KEY), and flip the BizyAirPlus action-bar toggle ON.

Gotchas

Setup is the same three-suspect story - missing key, toggle off, wrong Python (python -m pip show bizyair-cloudberry). The Pro-specific mistake is treating it as a "fix my prompt" button: it's a better model, not a better prompter. If your brief is vague, Pro will produce a confidently wrong image just as fast as the standard tier - and you'll pay the premium rate for it. Draft with the cheap node, go Pro when the brief is already tight.