☁️Dall_e text2Image

This is the direct version of DALL-E 3 access in comfyui_LLM_party. If you've come across the pack's dall_e_tool node, that one hands image generation to an LLM agent as a callable capability - the agent decides when to generate and writes its own prompt at call time. This node skips the agent entirely: you write the prompt yourself, right on the node, wire it in like any other image-generation step, and get a real image back.

How it works

prompt is your text, image_size/image_quality/style are DALL-E 3's own API parameters, not something this pack invented, and the node calls OpenAI's image endpoint directly.

The inputs and outputs that matter

• prompt (multiline) - your image description. Required.
• image_size (1024x1024 / 1792x1024 / 1024x1792, default square) - pick portrait or landscape if that suits your subject better than square.
• image_quality (standard / hd, default hd) - hd costs more per DALL-E 3's pricing; drop to standard if you're iterating and don't need the extra fidelity yet.
• style (vivid / natural, default natural) - vivid pushes toward more hyper-real, dramatic output; natural aims for more realistic, less stylized results.
• base_url (default https://api.openai.com/v1/) / api_key - standard OpenAI credentials; per the README, base_url can also point at a compatible relay as long as it still ends in /v1/.
• is_enable - the pack's standard bypass toggle.

Two outputs: img (IMAGE), the actual generated picture wired straight into any downstream ComfyUI image node, and img_url (STRING), the raw hosted URL DALL-E returned - useful for logging, a report string, or referencing the image without redownloading it.

How to install it

• ComfyUI Manager: search "comfyui_LLM_party", install, restart.
• Manual: cd ComfyUI/custom_nodes && git clone https://github.com/heshengtao/comfyui_LLM_party, then pip install -r requirements.txt from inside the pack folder using ComfyUI's own Python (python_embeded\python.exe -m pip install -r requirements.txt on portable Windows), then restart.

The real prerequisite is external: a real OpenAI API key with billing active. DALL-E 3 is billed per image, and there's no free tier.

Common issues & troubleshooting

Calls fail with an auth error. The node ships with a placeholder sk-XXXXX default in api_key - if that never got replaced with a real key, every call fails immediately. Separately, billing needs to be active on the account, not just a valid key.

Costs climb faster than expected. hd quality and non-square sizes both cost more per DALL-E 3's own pricing tiers. Iterate on standard and square, lock in hd and your final aspect ratio once the prompt is dialed in.

The pack fails to load at all. Separate from this node, comfyui_LLM_party has a real, recurring install issue where nodes fail to import after a Manager install, usually a dependency or CUDA/PyTorch mismatch. If openai_dall_e isn't showing up in your node browser, check the ComfyUI console for an import error before assuming anything's wrong with your DALL-E setup specifically.