BFL Image Generator (Pro 1.1/1.0)

The BFL Image Generator is the plainest node in this pack: prompt in, image out, via BFL's flux-pro-1.1 API. If you want Flux Pro quality without installing a single local model, this is the minimum viable version of that. It's also the node with a small built-in lie worth knowing about before you get confused.

The display name says "Pro 1.1/1.0", and there's an api_version dropdown - but it only offers 1.1. Version 1.0 was removed from the API in the pack's v3.0.0 rewrite, and the "1.0" in the name is leftover branding. More importantly, that removal made the steps and guidance inputs into ghosts: the code only sends them when api_version is 1.0, which you can't select anymore. The knobs are there, you can spin them, nothing happens. Change steps all you want - flux-pro-1.1 ignores them, because that endpoint does its own thing internally.

What it is

A node from charlyad142/ComfyUI_bfl_api_pro_nodes calling the flux-pro-1.1 endpoint. It's the pack's basic text-to-image: set dimensions, drop an optional reference image, go. There's a reason it's the least-searched node in the set - for most people the BFL Flux Ultra node in the same pack is the better spend, since Ultra is the "maximum quality" tier. This one is the plain, predictable, cheapest-flavor option.

How it works

Your prompt (and any image_prompt) is POSTed to https://api.bfl.ai/v1/flux-pro-1.1 with your API key in the x-key header. BFL returns a task ID, the node polls get_result, downloads the finished image from Azure Blob Storage, and outputs an IMAGE tensor plus a STRING status line (✓ or ✗ Error: ...). Failure gives you a red placeholder image and the error text in the string.

The inputs that matter

• prompt - required, multiline. Flux's T5 encoder likes full sentences, so describe naturally.
• width / height (512–2048, defaults 1024×768) - unlike the Flux 2 nodes, these are real and required here. Set your canvas.
• image_prompt - an optional reference image for img2img-style guidance. Everything else stays prompt-driven.
• seed - default is 30 (not -1!), so out of the box every run is reproducible. Set it to a random value when you actually want variety.
• steps / guidance - see above: accepted, ignored on 1.1. Don't waste time tuning them.

safety_tolerance (0–6) and output_format (jpeg/png) round out the required set, with x_key for your API key.

Installing it

ComfyUI Manager, search "ComfyUI BFL API Pro Nodes", or:

cd ComfyUI/custom_nodes
git clone https://github.com/charlyad142/ComfyUI_bfl_api_pro_nodes

Restart. No models, no requirements.txt - requests and Pillow, already in ComfyUI.

API key needed. Recommended: config.ini in the node's root folder:

[API]
X_KEY = your_api_key_here

That takes priority over the per-node x_key field. Keys from auth.bfl.ai.

Troubleshooting

Pack-standard failures: 401 bad key or no credits, 422 validation, 429 rate limiting, moderation messages from BFL's filters. The one genuinely node-specific trap is expecting steps/guidance to do anything - they won't, and that's by design of the 1.1 endpoint. If your results feel too plain, don't fight this node; reach for Flux Ultra or Flux 2 Flex instead, which actually expose the controls.