Concurrent Submit | Zhenzhen_gpt_image_1

Of all the OpenAI nodes in the Comfyui-zhenzhen pack, this is the simple one: prompt in, image out. GPT Image 1 is OpenAI's image-generation-and-editing model, and while the pack's edit node gives you masks and fidelity dials, this node keeps the surface small - quality, size, format, background, moderation, and how many images you want. It's the node you'd use for clean text-to-image work where you don't need surgical editing, or when you just want to throw a batch of prompts at OpenAI's best current image model.

It's still an API call, resold through the author's key - nothing runs locally, and you pay per image.

How the concurrent bit works

The plain Comfly_gpt_image_1 node blocks until OpenAI answers. The Submit variant runs it in the pack's shared image pool (30 workers) and returns a task immediately. Collect via Concurrent Collect Images (30): wire task_1 … and the collector waits for the batch, returning images in slot order plus a status string. failure_mode = fail-fast (default) or placeholder.

For "I need twenty concept images by lunch," this is the node - all twenty prompts fire in parallel, one collector run.

The inputs that matter

• prompt (required) - the description.
• model - gpt-image-1 or gpt-image-1.5.
• n - 1–10 images per call. Metered per image, so this multiplies your spend.
• quality - auto / high / medium / low. Auto is fine; high is where the credits disappear.
• size - auto or the three fixed presets (1024×1024, 1536×1024, 1024×1536).
• background - auto / transparent / opaque. Transparent is handy for asset work (and only makes sense on supported sizes).
• output_format - png (default) / jpeg / webp.
• moderation - auto or low. "Low" relaxes OpenAI's filter - expect the model to still refuse things it will refuse.

Output: one task (COMFLY_IMAGE_FUTURE) → Collector → images.

Installing it

Ships in Comfyui-zhenzhen:

• ComfyUI Manager → Install Custom Nodes → search "Comfyui-zhenzhen" → install → restart.
• Or cd ComfyUI/custom_nodes && git clone https://github.com/T8mars/Comfyui-zhenzhen and restart.

No model downloads. You need a paid token from the author's storefront (ai.t8star.org), in api_key.

Gotchas

The README's ChatGPT-node history is honest about the tiers: the default "official-reverse" token group is cheap but slow and unstable, while the premium groups are faster but cost more - and the two families of tokens aren't interchangeable between nodes. If you're getting slow generations, that's the default group, not your graph. A 500 here is usually OpenAI load - the author's advice is retry, and occasionally the model simply returns no image; check the shop's log for whether you were actually charged before rerunning.