Flux1-DedistilledMixTuned - v3.0 fp8
é¢é£ããŒã¯ãŒãïŒã¿ã°
æšå¥šãã©ã¡ãŒã¿
samplers
steps
resolution
vae
æšå¥šãã€ã¬ãŸãã©ã¡ãŒã¿
upscaler
ãã³ã
ããè¯ãããã³ããçè§£èœåã®ããã«GNER-T5-XXLã®äœ¿çšãæšå¥šããŸãã
è§£å床1024x1024以äžã®ç»åã«ã¯ãeuler/deisãšnormal/beta/simpleã®ãµã³ãã©ãŒã䜿çšããŠãã ããã
1024x1024ãã2048x2048ã®å€§åç»åã«ã¯ãddim/dpm_2/dpmpp_2m/heunpp2ãšddim_uniformãŸãã¯betaã¹ã±ãžã¥ãŒã©ã䜿çšããŠãã ããã
æã詳现ãåŸãã«ã¯dpmpp_2m+betaãæãèžè¡ç广ãåŸãã«ã¯heunpp2+ddim_uniformã䜿çšããŠãã ããã
åºæ¬çãªãµã³ãã©ãŒçµã¿åããã¯deis+simpleãŸãã¯euler+betaãããå€ãã®ãã€ãºãå«ãããªãšãŒã·ã§ã³ãšããŠddim/dpm_2/dpmpp_2+betaãŸãã¯sgm_uniformããããŸãã
35mmãAGAFãKodakãªã©ã®LUTã䜿ã£ãŠãã£ã«ã 广ã远å å¯èœã§ãã
ã¢ãã«ã¯å±€ç¶èåæè¡ã䜿ããèžçã®å¹²æžãé€å»ãLoRAéã¿æåºŠãåäžãããŠããŸãã
ã«ã¹ã¿ã ããŒãäžèŠã®ã·ã³ãã«ãªUNETã¯ãŒã¯ãããŒãå©çšã§ããGGUFçã䜿ãå Žåã¯UNET Loader(GGUF)ããŒãã䜿ã£ãŠãã ããã
ããŒãžã§ã³ã®ãã€ã©ã€ã
V3.0çã¯å šé¢çã«ã¢ããã°ã¬ãŒããããçŸåšã®Flux Devãã¡ã€ã³ãã¥ãŒãã³ã°ã¢ãã«ã®äžã§æããã©ã³ã¹ã®ãšããã¢ãã«ã§ãããLoRAäºææ§ããªã¢ãªãºã ãç»åå質ãèžè¡çåµé æ§ã«ãããŠFlux Proã¢ãã«ã«æãè¿ããã®ã§ãã
Fully upgraded Version 3.0, it may be the best model in the current Flux Dev fine-tuning models. Have the very good balance in model capabilities, LoRA compatibility, realism, image quality and artistic creativity closest to the Flux Pro model.
ã¯ãªãšã€ã¿ãŒã¹ãã³ãµãŒ
ããè¯ãããã³ããçè§£èœåã®ããã«GNER-T5-XXLã®äœ¿çšãæšå¥šããŸãã以äžããããŠã³ããŒãããŠãã ããïŒhttps://civitai.com/models/1888454 ãŸãã¯ç§ã®HFãªããžããªã
ãŸã Huggingface.co ã«ããããŸãã
GGUFããŒãžã§ã³ã®ã€ã³ã¹ããŒã«ãšå€æãµããŒãã«ã€ããŠã¯https://github.com/city96/ComfyUI-GGUFãã芧ãã ãããã¢ãã«å€æã¹ã¯ãªããã¯https://github.com/leejet/stable-diffusion.cppã«ãããŸãã
V3.0-Krea ããŒãžã§ã³ïŒ
Flux.1-Dev-Kreaã¢ãã«ã¯ãDevã¢ãã«ã®èžè¡çã¹ã¿ã€ã«ãšãªã¢ã«ãªåç衚çŸèœåãåäžãããŸããããããŒãã¬ãŒãã®é®®æããšçŸåŠã¯äœäžããç¹ã«ãªãªãžãã«Devã¢ãã«ã§ãã¬ãŒãã³ã°ããLoRAãšã®äºææ§ãéåžžã«äœãã§ããæ¬V3.0-Kreaã¯Kreaã¢ãã«ã®äž»ãªé·æãä¿æããç»åã®é®®æããšãªãªãžãã«Devã¢ãã«LoRAãšã®äºææ§ãæ¹åããŠããŸãããLoRAäºææ§ã®æ¹åã¯éå®çã§çæ³çã§ã¯ãããŸãããããŠã³ããŒãã¯æ éã«ããŠãã ããã
Flux.1-Dev-Krea has improved the artistic style and realistic photography ability of the Dev version model, but the clarity and aesthetics of portraits have weakened, especially with poor compatibility with Lora trained on the original Dev model. The V3.0-Krea retains the main features of the Krea model, improves image clarity, and enhances compatibility with the Lora, but the Lora compatibility is improved minimal and not ideal, which is a disappointing aspect of this version. Please download cautiously.
ããè¯ãããã³ããçè§£èœåã®ããã«T5-XXLã®ä»£ããã«GNER-T5-XXLã®äœ¿çšãæšå¥šããŸãã以äžããããŠã³ããŒãå¯èœã§ãïŒhttps://civitai.com/models/1888454 ãŸãã¯ç§ã®HFãªããžããªã
Recommended to use GNER-T5-XXL instead of T5-XXL for better prompt understanding capabilities, you can download it from https://civitai.com/models/1888454 or my HF Repo.

ããã€ãã®ãµã³ãã«ç»å (Some example image) :

ã¢ãã«ã®äœ¿ç𿹿³ïŒ
åºæ¬ã®çµã¿åããïŒdeis+simple / euler+betaãããŸããŸãªçµã¿åãããã詊ããã ããã
åºæ¬: deis+simple / euler+beta, You can try more different combinations.
ãŸã Huggingface.co ã«ããããŸãã
===========================================================================
V3.0-PAP ããŒãžã§ã³ïŒ
v3.0-PAPïŒããŒãã¬ãŒãããã³èžè¡åçã®åºåæé©åããŒã¹ã¢ãã«ãæ§å³ãå
ãšåœ±ãæ±æŽäººã®é¡åœ¢ã«ç¹åããŠæé©åãããé¡ã¢ãã«ã®æåºŠãšé©å¿æ§ãããã«åŒ·åã
ãªãªãžãã«ã®Devã¢ãã«ãšæ¯èŒããŠããã®ããŒãžã§ã³ã¯äººçš®ãšé¡åœ¢ã«ãããŠãããªã¢ã«ã§è±ãã§ããé¡ã¢ãã«ã®äŸç»åã¯ä»¥äžã®äœè ã®LoRAã¢ãã«ããã®ãã®ã§ããååã«æè¬ããŸãïŒããèäœæš©äŸµå®³ãããã°ãé£çµ¡ãã ãããçŽã¡ã«åé€ããŸãã
Portrait and Art Photography Optimization Base Model. The model has been specially optimized in composition, light and shadow, oriental face shape to further strengthen the sensitivity and adaptability of the face model.
Compared to the Flux.1 Dev original model, this version is more realistic and richer in terms of ethnicity and face shape, and the example drawings of the face model are from the LoRA model of the following authors, thanks in advance! If there is any infringement, it will be notified and deleted immediately.
https://civitai.com/user/el_fluppe
https://civitai.com/user/wolfcatz
https://civitai.com/user/seanwang1221
https://civitai.com/user/nawusijia
ã¢ãã«ã®ç°¡åãªäœ¿çšã¬ã€ãïŒ
åºæ¬æ§æïŒdeis+simple / euler+beta; ãã€ãºå€ãïŒddim/dpm_2/dpmpp_2+beta/beta57/sgm_uniform; 詳现è±å¯ã§æ³ååè±ãïŒheunpp2+ddim_uniform; æ¡å€§ïŒUltimateSDUpscale/TTP; ãã£ã«ã 广ïŒLUT远å (35mm/AGAF/Kodak); ãŸãã¯ããªãã®ç°å¢ã«åºã¥ããæé©ãªçµã¿åãããã¹ãããã¯20-30ã§ããã¯ãŒã¯ãããŒã¯äŸç»åãåç §ããŠãã ããã
Basic: deis+simple / euler+beta; More noise: ddim/dpm_2/dpmpp_2+beta/beta57/sgm_uniform; More detail, more imaginative: heunpp2+ddim_uniform; Upscaler: UltimateSDUpscale/TTP; Film effects: add LUT (35mm/AGAF/Kodak); Or the best combination of tests based on your own environment. steps 20-30ã The workflow is shown in the example POST image.
ãŸã Huggingface.co ã«ããããŸãã
é¢çœãé¡ã¢ãã«LoRAã³ã³ãããŒã«ãã¹ãäŸ (An interesting face model LoRA control sample)ïŒ
===========================================================================
DedistilledMixTuned Dev V3.0ïŒ
äžåœã®å¹²æ¯ã®èå¹Žã«æ§ãå€§å¹ ã¢ããã°ã¬ãŒãïŒ
V3.0ã¢ãã«ã¯å®å šã«ã¢ããã°ã¬ãŒããããFlux Dev埮調æŽã¢ãã«ã®äžã§æããã©ã³ã¹ã®åããã¢ãã«ã§ãããLoRAäºææ§ããªã¢ãªãºã ãç»åå質ãèžè¡çåµé æ§ãFlux Proã«æãè¿ãã¢ãã«ã§ãã(æ¯èŒè©äŸ¡ã容æã«ãããããæ¬ã¢ãã«ã®ã·ãŒãã¯ãªãªãžãã«Devã¢ãã«ãšåºæ¬çã«æŽåããŠããŸã)
å®å šã«ã¢ããã°ã¬ãŒããããVersion 3.0ã¯ãçŸåšã®Flux Devãã¡ã€ã³ãã¥ãŒãã³ã°ã¢ãã«ã®äžã§æè¯ã®ãã©ã³ã¹ãæã¡ãLoRAäºææ§ããªã¢ãªãºã ãç»åå質ãèžè¡çåµé æ§ãFlux Proã¢ãã«ã«æãè¿ãã§ãã(è©äŸ¡ãšæ¯èŒã®ãããã·ãŒãã¯ãªãªãžãã«Devã¢ãã«ãšã»ãŒåãã§ã)

V3.0ã¢ãã«äœ¿çšã¬ã€ãïŒ
ã¢ãã«ã¯å±€ç¶èåæè¡ãçšããŠèžçã®å¹²æžãé€å»ãããªãªãžãã«Flux.1 Devã¢ãã«ãšã®å®å šãªäºææ§ãæã¡ãããé«ãLoRAéã¿ã®æåºŠãæã¡ãŸãã1024x1024以äžã®è§£å床ã§ã¯ãeuler/deis + normal/beta/simpleçãæšå¥šãã1024ã2048ã®å€§ããè§£å床ç»åã«ã¯ddim/dpm_2/dpmpp_2m/heunpp2 + ddim_uniform/betaãæšå¥šãããŸãã
æã詳现ã匷ãã®ã¯dpmpp_2m+betaãæãèžè¡æ§ãé«ãã®ã¯heunpp2+ddim_uniformã§ãã
æšå¥šã¹ãããæ°ã¯KSamplerã§20-30ã§ããã¯ãŒã¯ãããŒã¯ä»¥äžã®ç»åãåç §ããŠãã ããïŒhttps://civitai.com/images/53432419
The model is fully compatible with the original Flux.1 Dev. Had removed the de-distillation interference, and has a higher sensitivity to LoRA weights. For 1024x1024 and below, euler/deis + normal/beta/simple, etc., 1024 - 2048 for large binning images, ddim/dpm_2/dpmpp_2m/heunpp2 + ddim_uniform/beta.
More details: dpmpp_2m + beta, More artistry: heunpp2 + ddim_uniform
Recommended: KSampler, steps 20-30. Workflow of the model pls ref: https://civitai.com/images/53432419
===========================================================================
DedistilledMixTuned Dev V2.0ïŒ
2025幎æ°å¹Žã®èŽãç©ïŒ1ãæä»¥äžãããŠéçºããv1.0ãããŒã¹ã«å šé¢çã«ã¢ããã°ã¬ãŒãããV2.0ã¯ãåçå質ã®ãªã¢ãªãºã ã§ããã£ããŒã«ã®è¡šçŸãåºåé床ãLoRAäºææ§ãå ãšåœ±ã®èª¿åã®æ°ãããã©ã³ã¹ãå®çŸããŠããŸãã
2025 New Year Gift! More than a month of training and fine-tuning, The V2.0 version has been upgraded based on v1.0, and has reached a better balance in detail reflection, drawing speed, LoRA compatibility, light and shadow harmony with photorealistic realism.
===========================================================================
DedistilledMixTuned Schnell V1.0ïŒ
Flux.1 SchnellããŒã¹ã®ã¢ãã«ã§ã4ïœ8ã¹ãããã®é«éåºåããªãªãžãã«Flux Schnellã®æ§å³ã¹ã¿ã€ã«ã«åŸããåªããããã³ãã埩å èœåããããç»åå質ã现éšããªã¢ãªãºã ãã¹ã¿ã€ã«å€æ§æ§ã®æè¯ãã©ã³ã¹ãåãããªãŒãã³ãœãŒã¹ã§åçšå©çšå¯èœãªSchnellã®ããŒã¹ã¢ãã«ã§ãã
Only 4 step, The Model may achieve to the best balance in terms of image quality, details, reality, and style diversity compare with other tuned of Flux.1 Schnell. and have a good ability of prompt following, good of the original Flux model style following.
FLUX.1-schnellãããŒã¹ã«ãLibreFLUXãããŒãžããComfyUIãBlock_Patcher_ComfyUIãComfyUI_essentialsãªã©ã®ããŒã«ã§ãã¡ã€ã³ãã¥ãŒãã³ã°ãæšå¥šã¹ãããæ°ã¯4ïœ8ã§ãéåžž4ã¹ãããã§ååãFlux.1 Schnellä»ã¢ãã«ã«æ¯ã¹ãŠå質ãšãªã¢ãªãºã ãå€§å¹ ã«åäžã
===========================================================================
DedistilledMixTuned Dev V1.0ïŒ
10ã¹ããã以å ã®é«éåºåFlux埮調æŽã¢ãã«ã®äžã§ããªãªãžãã«Flux.1 Devã¹ã¿ã€ã«ã«å¿ å®ã§ãåªããããã³ãã埩å èœåãæé«ã®åºåå質ãFlux.1 Devã¢ãã«ãè¶ ãã现éšãæã¡ãFlux.1 Proã«æãè¿ãããŒã¹ã¢ãã«ãããããŸããã
May be the Best Quality Step 6-10 Model, In some details, it surpasses the Flux.1 Dev model and approaches the Flux.1 Pro model. and have good ability of prompt following, good of the original Flux.1 Dev style following.
Flux-Fusion-V2ãããŒã¹ã«ãflux-dev-de-distillãããŒãžããComfyUIãBlock_Patcher_ComfyUIãComfyUI_essentialsãªã©ã®ããŒã«ã§ãã¡ã€ã³ãã¥ãŒãã³ã°ãæšå¥šã¹ãããæ°ã¯6ïœ10ãFlux.1ã®ä»ã®ã¢ãã«ã«æ¯ã¹ãŠå質ãå€§å¹ ã«åäžã

GGUF Q8_0 / Q5_1 /Q4_1éååã¢ãã«ãã¡ã€ã«ã¯ãã¹ãæžã¿ã§åæã«æäŸäžãéå°ãªéååã¯é«éã»é«ç²ŸåºŠã¢ãã«ã®å©ç¹ã倱ããããä»ã®éååã¯æäŸããŸãããå¿ èŠãªæ¹ã¯ä»¥äžã®æ¹æ³ã§FP8ã¢ãã«ãããŠã³ããŒããããèªèº«ã§éååããŠãã ããã
GGUF Q8_0 / Q5_1 /Q4_1 quantized model file, had tested, and uploaded the same time, over-quantization will lose the advantages of this high-speed and high-precision model, so no other quantization will be provided, you can download the FP8 model file and quantizate it according to the following tips.
===========================================================================
æšå¥šïŒ
UNETããŒãžã§ã³ïŒã¢ãã«åäœïŒã¯ããã¹ããšã³ã³ãŒããŒãšVAEãå¿ èŠã§ãã以äžã®CLIPãšããã¹ããšã³ã³ãŒããŒã¢ãã«äœ¿çšãæšå¥šãããã³ããèªå°ãæ¹åãããŸãïŒ
ãã³ã°CLIPïŒhttps://huggingface.co/zer0int/CLIP-GmP-ViT-L-14/blob/main/ViT-L-14-TEXT-detail-improved-hiT-GmP-TE-only-HF.safetensors
ããã¹ããšã³ã³ãŒããŒïŒhttps://huggingface.co/silveroxides/CLIP-Collection/blob/main/t5xxl_flan_latest-fp8_e4m3fn.safetensors
VAEïŒhttps://huggingface.co/black-forest-labs/FLUX.1-schnell/tree/main/vae
GGUFããŒãžã§ã³ïŒGGUFã¢ãã«å¯Ÿå¿ããŒãã®ã€ã³ã¹ããŒã«ãå¿ èŠãhttps://github.com/city96/ComfyUI-GGUF
ç°¡åãªã¯ãŒã¯ãããŒïŒéåžžã«ã·ã³ãã«ãªã¯ãŒã¯ãããŒã§ãä»ã®Comfyã«ã¹ã¿ã ããŒãã¯äžèŠã§ãïŒGGUFããŒãžã§ã³ã¯åž96æ°ã®UNET Loader(GGUF)ããŒãã䜿çšããŠãã ããïŒïŒ
===========================================================================
èžçã®èã£ãããæŽãæµããã¢ãã«ã®æ¬è³ªã«ååž°ããæãçŽç²ãªFluxé«å質ããŒã¹ã¢ãã«ã®æ§ç¯ã«æ³šåïŒ
Wash away the distillation and return to the original basic.
ã¢ãã«ãæ°ã«å ¥ã£ããããã²äœ¿çšç»åãããããæçš¿ããŠãã ãããããããšãããããŸãïŒ
If you feel the model is good for you, please post the image here, thanks a lot!
Thanks for:
https://huggingface.co/black-forest-labs/FLUX.1-devãFLUX.1 [dev]éåçšã©ã€ã»ã³ã¹äžã®éåžžã«åªãããªãŒãã³ãœãŒã¹T2Iã¢ãã«ã
https://huggingface.co/black-forest-labs/FLUX.1-schnellãApache-2.0ã©ã€ã»ã³ã¹ã®éåžžã«è¯ããªãŒãã³ãœãŒã¹T2Iã¢ãã«ã
https://huggingface.co/AnibaaalãFlux-Fusionã¯çŽ æŽãããããã¯ã¹ïŒãã¥ãŒã³ã¢ãã«ã
https://huggingface.co/nyanko7ãFlux-dev-de-distillã¯çŽ æŽãããå®éšçãããžã§ã¯ãã§ãïŒinference.pyã¹ã¯ãªããã«æè¬ã
https://huggingface.co/jimmycarter/LibreFLUXãç¡æã®ãã¹ãã«FLUXã¢ãã«ã§ãFLUX.1-schnellã®Apache 2.0ããŒãžã§ã³ã
https://huggingface.co/MonsterMMORPGãFurkanæ°ã¯Flux.1ã¢ãã«ã®ãã¹ãã»ãã¥ãŒãã³ã°ã³ãŒã¹ãå€ãå ±æãç¹ã«ãã¹ãã«ã¢ãã«ã®ç¹å¥ãã¹ãã«è²¢ç®ã
https://github.com/cubiq/Block_Patcher_ComfyUIãcubiqæ°ã®Fluxãããã¯ãããã£ãŒãµã³ãã©ãŒã§Flux.1ã®ãããã¯ãã©ã¡ãŒã¿ãç»åçæã«ã©ã圱é¿ããã倿°ã®ãã¹ããå¯èœã圌ã®ComfyUI_essentialsã«ã¯FluxBlocksBusterããŒããããããããã¯å€ã®èª¿æŽã容æãçŽ æŽãããä»äºã§ãïŒ
https://huggingface.co/twodgirlãã¢ãã«éååã¹ã¯ãªãããšãã¹ãããŒã¿ã»ãããå ±æã
https://huggingface.co/John6666ãã¢ãã«å€æã¹ã¯ãªããããã³ã¢ãã«ã³ã¬ã¯ã·ã§ã³ãå ±æã
https://github.com/city96/ComfyUI-GGUFãGGUFã¯ã©ã³ã¿ã€ãºã¢ãã«ã®ãã€ãã£ããµããŒãã
https://github.com/leejet/stable-diffusion.cppãçŽç²ãªC/C++ã§äœæãããGGUFã¢ãã«å€æã¹ã¯ãªãããæäŸã
泚æïŒGGUF Q5/Q4ãžã®å®¹æãªå€æã«ã¯ https://github.com/ruSauron/to-gguf-batã¹ã¯ãªããã䜿çšå¯èœã§ãããããããŠã³ããŒãããŠsd.exeãã¡ã€ã«ãšåããã£ã¬ã¯ããªã«çœ®ããfp8.safetensorsã¢ãã«ãã¡ã€ã«ããšã¯ã¹ãããŒã©ãŒã§batãã¡ã€ã«ãžãã©ãã°ãããšCMDãŠã£ã³ããŠãéããã¡ãã¥ãŒã«åŸã£ãŠå€æãå®è¡ãããŸãã
ã©ã€ã»ã³ã¹
ãã®éã¿ãã¡ã€ã«ã¯FLUX.1 [dev]éåçšã©ã€ã»ã³ã¹äžã«ãããŸãã
ã¢ãã«è©³çް
ã¢ãã«ã¿ã€ã
ããŒã¹ã¢ãã«
ã¢ãã«ããŒãžã§ã³
ã¢ãã«ããã·ã¥
äœæè
ãã£ã¹ã«ãã·ã§ã³
ã³ã¡ã³ããæ®ãã«ã¯ log in ããŠãã ããã
