Warning: Although these quants work perfectly with ComfyUI - I couldn't get them to work with Forge UI yet. Let me know if this changes. The original non-k quants can be found <a target="_blank" rel="ugc" href="https://civarchive.com/models/705444">HERE</a> which are verified working with Forge UI.[Note: Unzip the download to get the GGUF. Civit doesn't support it natively, hence this workaround]These are the K(_M) quants for HyperFlux 8-steps. The K quants are slightly more precise and performant than non-K quants. HyperFlux is a merge of Flux.D with the 8-step HyperSD LoRA from ByteDance - turned into GGUF. As a result, you get an ultra-memory efficient and fast DEV (CFG sensitive) model that generates fully denoised images with just 8 steps while consuming ~6.2 GB VRAM (for the Q4_0 quant).It can be used in ComfyUI with this <a target="_blank" rel="ugc" href="https://github.com/city96/ComfyUI-GGUF">custom node</a>. But I couldn't get these to work with Forge UI. See <a target="_blank" rel="ugc" href="https://github.com/lllyasviel/stable-diffusion-webui-forge/discussions/1050">https://github.com/lllyasviel/stable-diffusion-webui-forge/discussions/1050</a> for where to download the VAE, clip_l and t5xxl models.<h2 id="advantages-over-fastflux-and-other-dev-schnell-merges-wxnumnt3u">Advantages Over <a target="_blank" rel="ugc" href="https://civarchive.com/models/657607/gguf-fastflux-flux1-schnell-merged-with-flux1-dev">FastFlux</a> and Other Dev-Schnell Merges</h2><ul><li>Much better quality: you get much better quality and expressiveness at 8 steps compared to Schnell models like FastFlux</li><li>CFG/Guidance Sensitivity: Since this is a DEV model, unlike the Hybrid models, you get full (distilled) CFG sensitivity - i.e., you can control prompt sensitivity vs. creativity and softness vs. saturation.</li><li>Fully compatible with Dev LoRAs, better than the compatibility of Schnell models.</li><li>The only disadvantage: needs 8-step for best quality. But then, you'd probably try at least 8 steps for best results with Schnell anyway.</li></ul><h3 id="which-model-should-i-download-ufuv3ccuz">Which model should I download?</h3>[Current situation: Using the updated Comfy UI (GGUF node) I can run Q6_K on my 11GB 1080ti.]Download the one that fits in your VRAM. The additional inference cost is quite small if the model fits in the GPU. Size order is Q2 &lt; Q3 &lt; Q4 &lt; Q5 &lt; Q6. I wouldn't recommend Q2 and Q3 unless you absolutely cannot fit the model in memory.All the license terms associated with Flux.1 Dev apply.PS: Credit goes to ByteDance for the HyperSD Flux 8-steps LoRA which can be found at <a target="_blank" rel="ugc" href="https://huggingface.co/ByteDance/Hyper-SD/tree/main">https://huggingface.co/ByteDance/Hyper-SD/tree/main</a>

GGUF_K: HyperFlux 8-Steps K_M Quants

ggufKHyperflux8StepsK_q4KM.zip

Files with SHA256: 252635d4f3ed2f0349003a02d79be3561c9139ab9c65c8481b3b3de0df88b73c