A realistic base model block merged from schnellMODE and flux.1 dev.<h2 id="parameters-5ynxry6q5">Parameters</h2>FOR ALL MODELS - Use Euler / Beta, Euler / Normal, or Euler / Simple and 8 steps. Recommended resolutions: 1Mpix or 1.5Mpix, try 2Mpix, but your results will vary.Models with a 'g' in the version have text/prompt guidance layers. Use prompt guidance between 1 and 2.5 for the best images.<h2 id="notes-obr6nqd54">Notes</h2>Note: Somtimes the flux base model will push it's way through giving that flux sheen to skin and the "flux chin", most of the time you can try again and it'll go away.This is not based on schnell so is subject to the full flex dev license terms.NOTICE: NF4 in ComfyUI - I highly recommend this node: <a target="_blank" rel="ugc" href="https://github.com/DenkingOfficial/ComfyUI_UNet_bitsandbytes_NF4">DenkingOfficial/ComfyUI_UNet_bitsandbytes_NF4</a> You must install bitsandbytes version &gt;0.43.3 using pip <code>pip install bitsandbytes==0.45.2</code><h2 id="versioning-nmhgejkhc">Versioning</h2>v4.0g - Whoops, forgot my own version number system. Gave it more photographic understanding and detailv0.3g - Added guidance to allow guidance config to work.v0.3 - Much more stable, a lot more good seeds, better textv0.2 - Less unhinged, more good seeds, experimentation needed!v0.1 - is fairly unhinged, some good seeds, some bad.<h2 id="quality-n54vr6of4">Quality</h2>Quality (best first): FP16 &gt; Q8 &gt; FP8 &gt; Q6_K &gt; Q4_1 &gt; Q4_K &gt; NF4 &gt; FP4 &gt; Q3_K_S Speed (fastest first): FP8 - GGUF - FP16 1MPix1:1 - 1024 x 1024 4:3 - 1152 x 896 16:9 - 1344 x 7681.5MPix1:1 - 1280 x 1280 4:3 - 1440 x 1120 16:9 - 1680 x 960 21:9 - 1920 x 8002Mpix 1:1 - 1408 x 1408 4:3 - 1664 x 1216 16:9 - 1920 x 1088

devMODE - 8 steps - (FLUX .1 DEV)

With guidance layer for extra control using Flux Guidance in ComfyUI / Prompt Guidance in Forge

Files with SHA256: 8c9cd1ca3b0b8016d7f154522174d974349b747ef3fc2de46486898413732579