GPT img OAuth Generate Advanced
The fashion-spec generator, now with your ChatGPT subscription instead of an API meter
- reference_image
- image
- revised_prompt
Take GPT img API Generate Advanced - the node built for front/back product shots of garments, no model, no props, no logos - and remove the API key. That's this node. It's aimed at the same fashion-planning workflow: you describe the garment, it comes back as a clean e-commerce-style catalog image on a white background. But instead of billing an OpenAI API key per run, it authenticates through your ChatGPT/Codex login.
For anyone iterating on garment specs all day, this is the version worth keeping in the workflow. Design a suit, check the front/back render, tweak the collar, render again - subscription-metered instead of credit-metered.
How it works
Identical mechanism to the API Advanced node, through a local OAuth proxy instead of a direct keyed call. The node composes your four instruction fields - design_request, generation_instructions, reference_instructions, hard_constraints - into one structured prompt with an explicit rule that design_request wins on any conflict. On run it checks the proxy on 127.0.0.1:10531, auto-starts it via npx -y openai-oauth if needed, and streams the request to OpenAI's Responses API with the image_generation tool in generate mode.
The inputs that matter
design_request- the authoritative description, any language. Highest priority.generation_instructions- reusable output rules; default is the front/back catalog layout.hard_constraints- the negatives: no model, no mannequin, no text, no logo.reference_image- optional IMAGE; first frame drives layout, the rest are garment-design cues.model- OAuth list:gpt-5.4(default),gpt-5.4-mini,gpt-5.5.oauth_port/auto_start_oauth- proxy control; defaults are fine.
Outputs
image- the catalog render as a ComfyUI IMAGE (batched ifn> 1).revised_prompt- the rewritten prompt, worth inspecting because a lot of instruction gets compressed.
Install
cd ComfyUI/custom_nodes
git clone https://github.com/hub2vu/Comfyui-GPT-img-node.git GPT-img
Then Node.js plus a one-time login:
npx @openai/codex login
Restart ComfyUI. Manager may not have this pack yet (registry version is Pending), so manual clone it is. No Python dependencies.
Common issues
Same OAuth failure modes as the rest of the pack: missing Node.js, or a proxy that won't start - check custom_nodes/GPT-img/logs/openai-oauth.log and re-login. And remember the design request is authoritative, so the constraint fields fight a losing battle if you contradict yourself there. Keep n at 1 while you tune the instructions; each additional image is another full request through the proxy.
Inputs (16)
| Name | Type | Default | Description |
|---|---|---|---|
| design_request | STRING | Men's navy summer suit. | — |
| generation_instructions | STRING | Create a product-only apparel design image for fashion planning. Output format: Show exactly two views of the same garment in one image: - front view on the left - back view on the right The garment must be shown alone on a clean white or very light gray background. The result should look like a professional e-commerce product catalog image or a fashion technical presentation image. The front and back views must represent the same garment design. Use the same color, fabric, sleeve shape, collar type, hem shape, and overall silhouette in both views. | — |
| reference_instructions | STRING | Use the first reference image, if provided, as a layout reference only for the front/back side-by-side composition. Use the other reference images only for garment category, silhouette, color, material feel, fabric texture, collar, sleeve, button, pocket, seam, and hem details. Do not copy the exact product design, brand, logo, text, watermark, person, pose, background, or photography artifacts. If reference images contain a human model, ignore the person and use only the garment design cues. The user design request has priority over the reference images if there is any conflict. | — |
| hard_constraints | STRING | No human model. No mannequin. No hanger. No body parts. No face. No arms. No legs. No person wearing the garment. No lifestyle background. No studio props. No table. No folded garment. No flat lay. No extra accessories. No text labels. No logo. No watermark. | — |
| model | COMBO | gpt-5.4 | 3 options: gpt-5.4, gpt-5.4-mini, gpt-5.5 |
| quality | COMBO | medium | 3 options: low, medium, high |
| size | COMBO | 1024x1024 | 13 options: 1024x1024, 1536x1024, 1024x1536, 1360x1024, 1024x1360, 1824x1024, +7 |
| moderation | COMBO | low | 2 options: low, auto |
| n | INT | 11–8 | — |
| oauth_port | INT | 105311024–65535 | — |
| auto_start_oauth | BOOLEAN | true | — |
| timeout_sec | INT | 30030–3600 | — |
| system_prompt | STRING | You are an image generation assistant. Use the image_generation tool to create the requested image. Preserve the user's prompt, requested style, language, subject, and composition as closely as possible. Return image output, not explanatory text. | — |
| system_prompt_inputopt | STRING | — | |
| user_prompt_inputopt | STRING | — | |
| reference_imageopt | IMAGE | — |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| image | IMAGE | — |
| revised_prompt | STRING | — |