<h2 id="v1.7s-bhjvxkj2u">v1.7S</h2><ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/629880?modelVersionId=1253682">Moxie Flux 1 DS v1.5S</a> (previous version - high percentage)</li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/641214?modelVersionId=760338">Flux ArtFusion 4-steps v1.2</a></li></ul><h3 id="v1.7d-c7yr4zwfl"> v1.7D</h3><ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/629880?modelVersionId=1453489">Moxie Flux 1 DS v1.6</a> (previous version - high percentage)<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/694493/c4pacitor?modelVersionId=1700674">C4PACITOR d v3 alpha </a>(slightly higher percentage)</li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/833086?modelVersionId=1443922">Colossus Project Flux v5 AIO</a></li></ul></li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/83a92087-0c76-48ee-a210-c5e9abee1a45/width=525/83a92087-0c76-48ee-a210-c5e9abee1a45.jpeg" /><h3 id="v1.6d-vxe8xqfjv">v1.6D</h3><ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/629880">Moxie Flux1 DS v1.5D</a> (high percentage)</li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/941929?modelVersionId=1319871">DedistilledMixTuned v3.0 fp8</a></li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/713986b4-2a32-4720-9855-7a21c3c05bec/width=525/713986b4-2a32-4720-9855-7a21c3c05bec.jpeg" /><h2 id="v1.5s-7g5o17b2a">v1.5S</h2><ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/996504/shuttle-31-aesthetic?modelVersionId=1116641">Shuttle 3.1 Aesthetic v1.0</a> (high percentage)</li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/118111?modelVersionId=1045901">Copax Timeless v.FluxReal (fast)</a></li></ul><h3 id="v1.5d-m0ah8i06o">v1.5D</h3>1st merge<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/833086?modelVersionId=1159932">Colossus Project Flux v.42_AIO</a> (high percentage)</li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/141592?modelVersionId=992642">PixelWave v.FLUX1-dev 03</a></li></ul>2nd merge<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/694493/c4pacitor?modelVersionId=1123235">C4PACITOR v.d_v2 </a>(high percentage)</li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3506b12b-186f-4df7-872b-c64bd12eaec2/width=525/3506b12b-186f-4df7-872b-c64bd12eaec2.jpeg" /><h3 id="-tqusx9dfo"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/9a1a4faa-0973-479d-949b-a0868adfe5ea/width=525/9a1a4faa-0973-479d-949b-a0868adfe5ea.jpeg" /></h3><h3 id="1.4-4kdjpnj6r">1.4</h3>This is the 'new base' for <a target="_blank" rel="ugc" href="https://civarchive.com/models/781094/moxie-fusion-flux-f16-e5m2">Moxie Flux Fusion</a>, f16 e5m2 version. While this base, I find it has more 'failed' attempts (probably higher success at more steps, and playing with guidance. I keep it at 3-3.5 guidance, and 22 steps). However, the images are unique from other flux models, and it produces some great images. It had made a great pairing for <a target="_blank" rel="ugc" href="https://civarchive.com/models/781094?modelVersionId=985269">Moxie Fusion Flux v1.35</a>, giving it some good variety away from the 'stock' model. 1st merge<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/833086?modelVersionId=931988">Colossus Project Flux v1.0</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/652009?modelVersionId=959302">Flux_Realistic SaMay v2</a></li></ul>2nd merge<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/808669?modelVersionId=904292">Dev2Pro </a>(just a splash - I make Lora's with it, and sometimes use it as an upscaler)</li></ul><h3 id="1.3-6iyr3snc4">1.3</h3>This is the 'new base' for <a target="_blank" rel="ugc" href="https://civarchive.com/models/781094/moxie-fusion-flux-f16-e5m2">Moxie Flux Fusion</a>, but the f8 version. 1st merge<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/118111?modelVersionId=778112">Copax Timeless XL XPlus_1</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/720719?modelVersionId=806846">Project Gaia Flux1.D (Uncensored)</a></li></ul>2nd merge<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/645943?modelVersionId=722620">Flux Unchained by SCG</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/463163?modelVersionId=742904">The Araminta Experiment (SDXL+Flux) Flux1-A1</a></li></ul>3rd merge<ul><li>Merge first two merges</li></ul>4th merge<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/645943?modelVersionId=722620">Flux dev1 fp8 e4m3fn</a> (as the primary)</li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/691446">ByteDance Hyper-FLUX Acceleration LoRA 8-steps</a></li></ul><h3 id="flux1.d-and-flux1.s-8-step-v0of2hsj2"> Flux1.D &amp; Flux1.S 8-Step</h3>This is the absolute 'base' checkpoint of <a target="_blank" rel="ugc" href="https://civarchive.com/models/618692/flux?modelVersionId=691639">Flux Dev/Schnell</a> (UNET, fp8, e4m3fn), merged with <a target="_blank" rel="ugc" href="https://civarchive.com/models/691446?modelVersionId=774008">ByteDance Hyper-FLUX Acceleration LoRA</a>. It is merges to be a full checkpoint, with dual clips, and VAE included. I'm sure these are somewhere else, but I was unable to find them, so I am uploading them. I find the quality about neck in neck with the 'full' version, and I use it quite often (for testing) since, obviously, 8-step is much faster than 20. After trying to find another upload of this, but finally gave up and decided to upload it. <s> </s> v1.1This is 50/50 merger of <a target="_blank" rel="ugc" href="https://civarchive.com/user/AbdallahAlswa80">AbdallahAlswa80</a>'s 2 speed models, <a target="_blank" rel="ugc" href="https://civarchive.com/models/629858?modelVersionId=704188">Speed FP8 (e4m3fn) D</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/629858?modelVersionId=706389">Speed FP8 (e5m2) S</a>. Produces images in 4 steps. See sample image metadata for specifics.<h3 id="v1.0-fhwydfcyz">v1.0</h3>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.<a target="_blank" rel="ugc" href="https://civarchive.com/user/Adel_AI">Adel_AI</a> 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 <a target="_blank" rel="ugc" href="https://civarchive.com/models/627377/moxie1776s-flux-workflows">here</a>.) This is simply a <a target="_blank" rel="ugc" href="https://civarchive.com/models/629880?modelVersionId=704214">50/50 merger</a> of the two <a target="_blank" rel="ugc" href="https://civarchive.com/user/Adel_AI">Adel_AI</a> Flux1 checkpoints, <a target="_blank" rel="ugc" href="https://civarchive.com/models/623525/flux1d-16gb?modelVersionId=697065">Flux1.D 16Gb</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/623224/flux1s-16gb?modelVersionId=696732">Flux1.S 16Gb</a>. Below is a copy and paste from his page that gives further info.

Moxie Flux1 DS

1.231.21 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 <a rel="ugc" href="https://civitai.com/models/629880?modelVersionId=835532">Moxie Flux 1 v1.21</a> with in <a target="_blank" rel="ugc" href="https://civitai.com/models/463163">The Araminta Experiment Fv1</a>, and while Araminta has adult content, it has been much easier to 'clothed' models. For 1.23, I merged in <a target="_blank" rel="ugc" href="https://civitai.com/models/340248">Sinfully Stylish (dramatic lighting) v.02</a>, and <a target="_blank" rel="ugc" href="https://civitai.com/models/411088">Detailed Perfection style v1.0</a>. I also lightened the applied lower weights to my own merged Lora's in the earlier version along the way (those have not been uploaded, just fixed in this base).

Files with SHA256: 2a057a93ef1ca2b1a7cc5dee4bae49dd33b7148255eff6183674a29218f2aaf1