<h1 id="generate-high-quality-images-in-20-seconds">Generate high-quality images in 20 seconds</h1><hr />Please post the first picture directly here in the Model Gallery. This is the best way to support creators :) <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f9aaeace-704a-4278-8110-bf82bf5bcfdf/width=525/f9aaeace-704a-4278-8110-bf82bf5bcfdf.jpeg" />Hello ♥<hr /><h3 id="1)-download-the-flux.1-dev-(nunchaku-version)">1) Download the Flux.1-dev (nunchaku version)</h3><a target="_blank" rel="ugc" href="https://civarchive.com/models/1785647?modelVersionId=2020910"> Flux Dev (Nunchaku) : </a><a target="_blank" rel="ugc" href="https://civarchive.com/models/1831757">text to img</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1785647?modelVersionId=2020910">https://civarchive.com/models/1785647?modelVersionId=2020910 </a><a target="_blank" rel="ugc" href="https://civarchive.com/models/1831757">Flux Krea (Nunchaku) : text to img</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1831757">https://civarchive.com/models/1831757</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1785647?modelVersionId=2022720">Flux Kontext (Nunchaku) : img to img</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1785647?modelVersionId=2022720">https://civarchive.com/models/1785647?modelVersionId=2022720/</a><hr /><h3 id="2)-download-my-workflow-for-lessflux.1-(nunchaku-version)greater">2) Download my workflow for &lt;Flux.1 (nunchaku version)&gt;</h3><a target="_blank" rel="ugc" href="https://civarchive.com/models/1831687?modelVersionId=2072850">Flux Dev / Krea 20 sec Workflow v1.0</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1831687?modelVersionId=2107254">Flux Dev / Krea 20 sec Workflow v2.0 (New)</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1831687?modelVersionId=2107703">Flux Kontext 20 sec workflow v1.0</a>32GB ram , RTX 4070 12GB vramModel: svdq-int4_r32-flux.1-krea-dev (Get professional portrait without Lora) Model: svdq-int4_r32-flux.1-dev (Get better img with Lora //anime, NSFW..) text_encoder &gt;path : ....\ComfyUI\models\text_encoders\ text_encoders 1 : t5xxl_fp8_e4m3fn_scaled  (same with Flux Dev) text_encoders 2 : clip_l Lora x5 (Any Flux Dev Loar you Like)Generate info: 22 step, 640x896Generate time = 16 - 25 sec<h2 id="lora-i-use-in-this-workflow">Lora I use in this workflow</h2><a target="_blank" rel="ugc" href="https://civarchive.com/collections/11428250">https://civarchive.com/collections/11428250 (check my Flux Lora collection)</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/685874/flux-detailer">https://civarchive.com/models/685874/flux-detailer</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1166973/aesthetic-details">https://civarchive.com/models/1166973/aesthetic-details</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/643886/flux-faetastic-details">https://civarchive.com/models/643886/flux-faetastic-details</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1470162/midjourney-v7-meets-flux-illustrious">https://civarchive.com/models/1470162/midjourney-v7-meets-flux-illustrious</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1371216/realanime">https://civarchive.com/models/1371216/realanime</a> path : ....\ComfyUI\models\loras\<hr /><h3 id="3)-open-comfyui-and-load-the-workflow-enjoy">3) open ComfyUI and load the workflow, enjoy</h3>All generate by Flux.1-krea-dev -&gt; (svdq-int4_r32-flux.1-krea-dev) in 20sec@ 1 photo<h3 id="workflow-name-:-1-nunchaku-flux.1-krea-dev(t2i)-5lora.json">workflow name : <a target="_blank" rel="ugc" href="https://civarchive.com/api/download/models/2072850?type=Archive&amp;format=Other">1-nunchaku-flux.1-Krea-dev(T2I) 5Lora.json</a></h3><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/c1619349-8ca3-4097-a2ae-d97868d2d2e2/width=525/c1619349-8ca3-4097-a2ae-d97868d2d2e2.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/35eaa7df-a048-4304-bd01-d9c6e7d9f242/width=525/35eaa7df-a048-4304-bd01-d9c6e7d9f242.jpeg" /><h3 id="workflow-name-:-1-nunchaku-flux.1-dev(t2i)-10lora-pro.json-v2.0">workflow name : <a target="_blank" rel="ugc" href="https://civarchive.com/api/download/models/2107254?type=Archive&amp;format=Other">1-nunchaku-flux.1-dev(T2I) 10Lora-Pro.json v2.0</a></h3><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/687959b4-1886-4ca9-b5c7-8f2e15469f6b/width=525/687959b4-1886-4ca9-b5c7-8f2e15469f6b.jpeg" /><h3 id="workflow-name-:-1-nunchaku-flux.1-kontext(i2i)_v1.json">workflow name : <a target="_blank" rel="ugc" href="https://civarchive.com/models/1831687?modelVersionId=2107703">1-nunchaku-flux.1-Kontext(I2I)_v1.json</a></h3><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b971170a-1664-4526-89ed-6d4d017843de/width=525/b971170a-1664-4526-89ed-6d4d017843de.jpeg" /><h3 id="all-generate-by-flux.1-krea-dev-greater-(svdq-int4_r32-flux.1-krea-dev)-in-20sec@-1-photo">All generate by Flux.1-krea-dev -&gt; (svdq-int4_r32-flux.1-krea-dev) in 20sec@ 1 photo</h3><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f9275127-9ee1-4416-b88b-7fc58c2ab946/width=525/f9275127-9ee1-4416-b88b-7fc58c2ab946.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/4515f1b2-9e71-4c5d-8783-4a14a62d809b/width=525/4515f1b2-9e71-4c5d-8783-4a14a62d809b.jpeg" />

Flux.1-dev / Krea / Kontext (nunchaku) My 20sec Workflow

Flux.1-dev (nunchaku version) Flux.1-krea-dev (nunchaku version)

Workflow-flux1-Krea-dev-nunchaku_(text2img)5Lora.zip

flux1KreaDevNunchakuMy_v10.zip

Flux Dev v1.0

Kontext v1.0

Flux Dev v2.0

About Flux.1-krea-dev <a target="_blank" rel="ugc" href="https://docs.comfy.org/tutorials/flux/flux1-krea-dev">https://docs.comfy.org/tutorials/flux/flux1-krea-dev</a>About Flux.1-krea-dev (nunchaku version) <a target="_blank" rel="ugc" href="https://huggingface.co/nunchaku-tech/nunchaku-flux.1-krea-dev">https://huggingface.co/nunchaku-tech/nunchaku-flux.1-krea-dev</a> Discover the cutting-edge power of nunchaku-flux.1-krea-dev, the latest advance in text-to-image AI models! Built on the robust FLUX.1-dev base, this model delivers stunning, high-quality image generation with optimized efficiency through innovative 4-bit quantization technology.With the included svdq-fp4_r32-flux.1-krea-dev.safetensors file, you can experience blazing fast inference speeds on Nvidia GPUs while maintaining excellent image fidelity. Whether you’re a digital artist, developer, or AI enthusiast, this model empowers you to create rich visuals effortlessly.Download now to unlock:<ul><li>Flux Krea is fully compatible with LoRAs designed for Flux Dev.</li><li>State-of-the-art image synthesis tuned for creative flexibility</li><li>Efficient performance optimized for Blackwell GPU architectures</li><li>Robust open-source foundation with cutting edge quantization research backing</li><li>Ready integration with Diffusers and ComfyUI for seamless workflow</li></ul>Elevate your AI-generated art and media projects with Elevate your AI-generated art and media projects with nunchaku-flux.1-krea-dev — innovation meets artistry in every pixel. Flux.1-krea-dev Release date : 2025.7.31<hr />Requirement :<ul><li>Minimum VRAM requirement is about 6 GB vram (such as RTX 2060 or 3060)</li><li>Recommended VRAM is 10 GB or more for better performance and stability</li><li>Settings like CPU offload and smaller batch sizes help when VRAM is limited</li><li>Minimum system-RAM requirement is about 16 GB Recommended system-ram: 32 GB</li></ul><hr /><h3 id="1)-download-the-flux.1-krea-dev-(nunchaku-version)">1) Download the Flux.1-krea-dev (nunchaku version)</h3><a target="_blank" rel="ugc" href="https://huggingface.co/nunchaku-tech/nunchaku-flux.1-krea-dev/tree/main">https://huggingface.co/nunchaku-tech/nunchaku-flux.1-krea-dev/tree/main</a> ---- Nvidia RTX50 svdq-fp4_r32-flux.1-krea-dev.safetensors (plz visit huggingface.co find <a target="_blank" rel="ugc" href="https://huggingface.co/nunchaku-tech/nunchaku-flux.1-krea-dev/tree/main">download</a>)---- Nvidia RTX 20 / 30 / 40 <a target="_blank" rel="ugc" href="https://civarchive.com/api/download/models/2072913?type=Model&amp;format=SafeTensor&amp;size=full&amp;fp=nf4">svdq-int4_r32-flux.1-krea-dev.safetensors (Click me = Download now)</a> This File has uploaded at this post --- path: \ComfyUI\models\diffusion_models\svdq-int4_r32-flux.1-krea-dev.safetensors--- 2 advanced text_encoders are recommended to make it easier for you to create masterpiecestext_encoders 1 : t5xxl_fp8_e4m3fn_scaled  (same with Flux Dev) text_encoders 2 : clip_l (you can just use this to replace HiDream) <s>text_encoders 3 : </s><a target="_blank" rel="ugc" href="https://civarchive.com/models/1805024?modelVersionId=2042711"><s>Long CLIP - Distilled for Flux (Download) (471.71MB)</s></a><s> optional !!!</s><a target="_blank" rel="ugc" href="https://civarchive.com/models/1805024?modelVersionId=2042711"> </a> path : \ComfyUI\models\text_encoders\<hr /><h3 id="2)-download-my-workflow-for-lessflux.1-(nunchaku-version)greater-20sec-1-photo-workflow">2) Download my workflow for &lt;Flux.1 (nunchaku version)&gt; { 20sec 1 photo Workflow }</h3><a target="_blank" rel="ugc" href="https://civarchive.com/models/1831687?modelVersionId=2072850">Flux Dev / Krea 20 sec Workflow v1.0 (2025.08.03)</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1831687?modelVersionId=2107254">Flux Dev / Krea 20 sec Workflow v2.0 (New 2025.08.13)</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1831687?modelVersionId=2107703">Flux Kontext 20 sec workflow v1.0 (2025.08.13)</a><hr /><h3 id="workflow-name-:-1-nunchaku-flux.1-dev(t2i)-10lora-pro.json-v2.0-(with-prompt-chatgpt-5.0)">workflow name : <a target="_blank" rel="ugc" href="https://civarchive.com/models/1831687?modelVersionId=2107254">1-nunchaku-flux.1-dev(T2I) 10Lora-Pro.json v2.0 (with prompt ChatGPT 5.0)</a></h3><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/687959b4-1886-4ca9-b5c7-8f2e15469f6b/width=525/687959b4-1886-4ca9-b5c7-8f2e15469f6b.jpeg" /><hr /><h3 id="workflow-name-:-1-nunchaku-flux.1-kontext(i2i)_v1.json">workflow name : <a target="_blank" rel="ugc" href="https://civarchive.com/models/1831687?modelVersionId=2107703">1-nunchaku-flux.1-Kontext(I2I)_v1.json</a></h3><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b971170a-1664-4526-89ed-6d4d017843de/width=525/b971170a-1664-4526-89ed-6d4d017843de.jpeg" /><hr />workflow name : <a target="_blank" rel="ugc" href="https://civarchive.com/models/1831687?modelVersionId=2072850">1-nunchaku-flux.1-Krea-dev(T2I) 5Lora.json (Old version but simple)</a><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/35eaa7df-a048-4304-bd01-d9c6e7d9f242/width=525/35eaa7df-a048-4304-bd01-d9c6e7d9f242.jpeg" /><hr /><h1 id="lora-i-use-in-this-workflow">Lora I use in this workflow</h1><a target="_blank" rel="ugc" href="https://civarchive.com/collections/11428250">https://civarchive.com/collections/11428250 (check my Flux Lora collection)</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/685874/flux-detailer">https://civarchive.com/models/685874/flux-detailer</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1166973/aesthetic-details">https://civarchive.com/models/1166973/aesthetic-details</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/643886/flux-faetastic-details">https://civarchive.com/models/643886/flux-faetastic-details</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1470162/midjourney-v7-meets-flux-illustrious">https://civarchive.com/models/1470162/midjourney-v7-meets-flux-illustrious</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/1371216/realanime">https://civarchive.com/models/1371216/realanime</a> path : ....\ComfyUI\models\loras\--- --- --- --- --- --- --- --- --- --- --- --- advanced text_encoders are recommended to make it easier for you to create masterpiecestext_encoders 1 : t5xxl_fp8_e4m3fn_scaled  (same with Flux Dev) text_encoders 2 : <a target="_blank" rel="ugc" href="https://civarchive.com/models/1805024?modelVersionId=2042711">HiDream CLIP - Distilled for Flux (Download) (471.71MB)</a> optional !!! text_encoders 3 : clip_l (You can use Flux Dev - text_encoders) path : ....\ComfyUI\models\text_encoders\<hr /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/4515f1b2-9e71-4c5d-8783-4a14a62d809b/width=525/4515f1b2-9e71-4c5d-8783-4a14a62d809b.jpeg" />

Flux krea dev (nunchaku svdq-fp4) Base-Model

Flux.1-krea-dev (nunchaku fp4) easier for you to create masterpieces artworks<ul><li>Minimum VRAM requirement 6 GB +</li><li>System-ram 16 GB +</li></ul>

Files with SHA256: 588d246f646c0dfa042326ad4c2ac4b2db75dedcccd4edccd8e1d72b0c9d9c52