1.23

1.21 simply lacked the depth, and, it was too likely to produce images that were a lot more risky than I was okay with (I'm sure you can fill in any blanks on your own). For version 1.22, I merged in The Araminta Experiment Fv1, and while it has adult content, it has been much easier to 'cloth' models. For 1.23, I merged in Sinfully Stylish (dramatic lighting) v.02, and Detailed Perfection style v1.0. I also lightened the applied weights my own Lora's in the earlier version along the way.

v1.21

This is the 'non-hyper' version of v1.2 (or as close as I could get). With this new base, I merged the same Lora's into it as I did w/ the 8 step and 4 step 1.2 versions. The base is Flux Unchained by SCG v1.1, and Copax TimelessXL vXPlus_1. Then I trained it by merging in some custom Lora's.

v1.2

Both 4 and 8 step are merged with the same Lora's. 8 step is D while 4 step is S.
8step

This version is a mix between a merge and a hybrid. I merged a couple of chekpoints together, Jib Mix Flux fp8 w/ Hyperlora, and Flux Unchained by SCG HyFU Hyper v1.0. Then I merged in some of my Lora's to come up with this version.

4step

This version is a mix between a merge and a hybrid. I merged a couple of chekpoints together, ArtFusion Flux CCA 4 Steps v1.0 v1.0, and Flux Unchained by SCG SchnFU v1.3 4 step. Then I merged in some of my Lora's to come up with this version.

v1.1

This is 50/50 merger of AbdallahAlswa80's 2 speed models, Speed FP8 (e4m3fn) D and Speed FP8 (e5m2) S. Produces images in 4 steps. See sample image metadata for specifics.

v1.0

There are other Flux.1 D and S merges, but none of them are an actually 'stand alone' checkpoint's. Hence, my motivation for this. Also, this has been tested on a 4060 16GB and 3060 12GB with no issues.

Adel_AI already created a couple of checkpoints that greatly simplify using Flux where you only have to download 1 file, and place it in your checkpoint folder for ComfyUI (reference my workflow here.)

This is simply a 50/50 merger of the two Adel_AI Flux1 checkpoints, Flux1.D 16Gb and Flux1.S 16Gb. Below is a copy and paste from his page that gives further info.

Flux.1 S [Shnell] is a 12 billion parameter T2I model developed by Black Forest Labs (https://blackforestlabs.ai), allowing to generate high-quality images in only 1 to 4 steps.

Safetensors with baked T5 fp8, Clip L and VAE
Put the model in the checkpoint directory of ComfyUI.