FUBUWA MIX

*Showcase images include embedded workflow and are post-processed with sharpening and noise grain.In other words, your results may vary depending on your workflow.Recommended settings are below.Version featuresRather than improvements, each version has their own strengthsv1.1 - Most realistic 2nd most realisticv1.2 - Semi realistic - sometimes include outlinev1.33 - Slightly less realistic than 1.1 in exchange for better lighting/composition/prompt adherence.v1.34 - v1.33 + slightly more detail in backgroundv2.4 - Most realisticv3.1 - Most balancedRecommendationFrom realistic to semi realistic = v2.4 > v1.1 > v3.1 > v1.34/1.33 > 1.2Personal favorite = v3.1 / v2.4 Version notesv3.1Most balanced version - combining strengths from previous versionsRealistic skin texture, good contrast/lighting/colour, better background detailsGood face proportionsGood prompt adherenceAesthetic more illustrative (Closer to v1.34 than v2.4)v2.4Focus more on realismMore realistic skin textureShould have better prompt adherenceAesthetic closer towards realistic photography than illustrative (Compared to v.1.0 variants)v1.34Should have slightly better details to the background.v1.33Focus on keeping realism while improving composition, background and lighting compared to v1.1.Less creative/aesthetic than 1.2 in terms of background.v1.2More focus towards semi realism *includes outline at certain timesBetter lighting, composition, background, aestheticStructure feels more stable compared to v1.1Model focus3DCG features with more detailed skin texture. (Heavily dependent on heun sampler, not recommended if using web generator or if you don't know what you're doing)Face proportions closer to real life. Less anime like.Strength/QuirksFacial features can respond to promptsWeaknessNot suitable for male because of female face modifier.Will do beauty mark/mole under eye (Apparent in lower cfg; use negative prompt)Skin pores can look too smooth/uniform/unnatural texture depending on sampler/scheduler.Backgrounds (res_multistep + normal with high steps 50-100 - has chance to resolve scale issues but might introduce artifacts) (better in version 1.2 and above)Recommended settings(can see workflow via uploaded images)positive promptscan change the face structure/maturenessmilf, adultmakeup/beauty enhancements(dark maroon eyeshadow under eye, maroon blush makeup, eyeliner:1), (doe eyes makeup:0.5), shiny skin, (perfect hands:0.9)realism tagssubsurface scattering, photorealistic, realistic, hyperrealisticinitial samplersa_solver_pece + sgm_uniformsweet spot: steps 30-40, cfg 6-7 @ 1024x1280 pixelsmore realistic at the cost of contrast/dark/warmer image: steps 30-40, cfg 4-5provides the most stable structure, details and subtle noise at reasonable speed. handles cfg burn not bad as well. use cfg higher than 7 for more contrast.dpm_adaptive + exponentialsteps doesn't matter since adaptive, cfg 4-10slower but very reliable without much errors. however compared to sa_solver_pece, it is smoother/no noise and lack small details.dpmpp_sde + sgm_uniform/exponential/sgm_uniformsteps 30, cfg 4-7feels slightly faster than the two above with reasonable structure.res_multistep + sgm_uniform/normal/simplesteps 30-100, cfg 4-6fast, reliable but slightly boring/simplistic resultsres_multistep is fast and converges at high steps, fixes the structural problems 90% of the time.upscale/hiresfixupscale modelanyrealesrgan_x2 (only because it's fast - don't think it makes a big difference which ever used)heun + normal/ddim_uniform(second choice)steps 15-20, cfg 3-4, denoise 0.35-0.55more random noise pattern with cleaner details compared to 2m_sde_heun karrasdpmpp_2m_sde_heun + karrassteps 20-30, cfg 3-5, denoise 0.35-0.55slightly faster than above, noise pattern more uniform compared to heunfacedetailerheun + normalsteps 10, cfg 2.5-6, denoise 0.25-0.4dpmpp_2m_sde_heun + karrassteps 10-15, cfg 2.5-6, denoise 0.25-0.4mole under eye might appear if low cfgRemarkssa_solver_pece + sgm_uniform is my new favourite combinationusual euler/2m/3m + sgm uniform/normal/karass should work thoughi prefer heun as it produces natural noise/texture at lower step values and loses it at higher (recommended schedulers: normal/ddim_uniform/ays+)for prompt adherence: higher resolution can have lower cfg, lower resolution need higher cfgpersonal frequently used resolution before upscale: 896x1120 1024x1280 or 1152×1440 (more pixels = more space for details like hands)post-processing nodes with sharpening, bloom, grain, tone curves for more texture in final image. (see workflow in images)