BigLove is a merge of Pony, bigASP and other SDLX models. The Pony releases of Big Love are a hybrid. They produces different images than both Pony & SDXL-based models. Pony tags are understood, but interprets them a bit differently. Some images lean more in the one or other direction depending on how you prompt. It works with Pony as well as normal SDXL loras. It can turn anime/cartoon into photorealistic images in img2img.

Big Love XL is a combination of 5 different training branches of SDXL: original SDXL, bigASP, RealVis, Anteros and Pony. With a bit of Pony in it, some Pony loras (poses, characters) work too. It outputs a more photorealistic & creative look than the Pony versions and is can also do paintings & cartoon. It adds a less strong texture by default, but you can increase it by adding texture, skin pores to the prompt & upscaling with up to 12 Lightning/DMD2 steps.

For more prompts see the other Big Love model page!

txt2img Settings:

1. DPM++ SDE, Normal/SGM Unifom (Pony), Karras (XL), 20-30 steps, cfg 3-10, clip skip 2, 832x1216 or 1024x1536

For faster generation 2M SDE/3M SDE works too, but lower quality with the Pony version.

Higher Quality:

2. For more contrast add the SDXL Lightning 8-step lora with DPM++ SDE, SGM Uniform, 8 steps, cfg 1.5-3 (Pony), 1-2.5 (XL), clip skip 2, 832x1216 or 1024x1536

3. For a more natural look use the DMD2 lora with LCM Exponential, 8 steps, cfg 1-2 (with lora weight 1-0.75), clip skip 2, 832x1216 or 1024x1536

Upscaling:

1. txt2img with Lightning 8-step lora: Hires fix, 4x_NMKD-Superscale-SP_178000_G upscaler, 3 hires steps, 0.4 denoise, 1.5x, hires cfg 2, clip skip 2

Higher Quality:

2. img2img with Lightning 8-step lora: same prompt, DPM++ SDE, SGM Uniform, 1.5x (XL), 2x (Pony), 3 steps (ComfyUI), 6-8 steps (Pony), 8-12 steps (XL), cfg 2, 0.3-0.5 denoise, clip skip 2

3. img2img with DMD2 lora: same prompt, LCM Exponential, 1.5x, 8-12 steps, cfg 1, 0.3-0.5 denoise, clip skip 2

Please note: Unlike txt2img with Hires fix, img2img offers a seed, so you get multiple image versions to choose from. img2img upscaling also produces better quality in A1111/Forge.

Strengths:

XL: Great fine details, very natural looking photo-like images, lovely faces, creative poses, can also do paintings and cartoon

Pony: Probably the most photorealistic Pony model, very good anatomy knowledge, huge face variety for a Pony model, very short prompts are quite effective & photorealistic

Weaknesses:

XL: Some prompts produce softer skin, multi-person images can have strange anatomy

Pony: Needs rather precise prompts & takes prompt rather literally, still a bit of the Pony look depending on used prompt, it can switch between fully photorealistic and semi-realistic

XL Prompting Advice:

SDXL prompts & prompts from Acorn Is Boning & Lustify work with it. Pony prompts too, but they create a more photorealistic look. Natural language prompts tend to create artistic glamour images with less skin detail while normal tags do more photo-style images. Big Love XL gives individual words more attention, so you don't need to rework prompts like Big Love Pony requires sometimes.

XL1 has a tendancy towards cuteness and bright skin. If you get too much of it, remove words like cute, sweet, pale/fair skin & flash from the prompt. It also creates realistical analog photos with reduced image quality. Sometimes better to remove terms like analog or grain to increase quality. Decrease cfg to get a more natural look.

Pony Prompting Advice:

Pony prompts & prompts from bigASP & Lustify work with it. Score tags and fill words like masterpiece or perfect skin have as good as no effect usually. A negative prompt is often unnecessary unless you want to make something vanish or avoid underage, but it can also change the style of the image.

Pure booru tags and some other samplers may produce Pony style. Photographic terms or a simple prompts ensure a photo look. Sometimes it refuses to produce medium shots, portraits or close-ups unless you weigh the term heavily or remove feet, shoes, high heels, boots etc. from the prompt. wavy hair sometimes does not look so great.

If it refuses to do ****, try simplifying the prompt. If you want to do a portrait, don't describe surroundings too much. If you want to do ****, describe the action mainly and don't focus on the people or surroundings in the prompt. The largest part of the prompt should be about the main subject of the image.

If a prompt does not produce the intended image, you often only need to modify it a bit to make it work perfectly.

Using Character Loras:

There were some complaints about them not working, but the problem was not the model itself so far. So here are some tips: * Try Pony as well as SDXL character loras * Use a lora weight up to 1.5 if it does not reduce quality * If it is a well known character, add the name in the prompt and also give it a higher weight if necessary. * Check your negative prompt for problems. Delete it if necessary and rewrite it from scratch. * Render a lot of images to be able to pick those with the best resemblance.

Higher Ouput Quality:

You can highly improve the images generated with this model by using img2img, upscaling, detailing etc. The Lightning lora enhances the look and sharpness despite only 8 steps. Alternatively, the DMD2 lora produces a sharp, but more natural look with less contrast. Also check out the Cuteness lora, which nicely enhances DMD2 images.

Merged Models:

XL2: TamePony 2.5, bigAsp 2, Lustify Engame, Acorn is Boning XL v1, Acorn is Boning XL v2, NatVis: Natural Vision 2, Anteros XXXL 1.0, XXX-Ray 1.0, Candidness2 lora, Realness lora, RealCuteness lora, Beauty lora

XL1: TamePony 2.5, bigAsp 2, Lustify Engame, Acorn is Boning XL v2

Pony2: TamePony 2.5, bigAsp 2, Lustify 4, URPM PonyXL-Hybrid 1

Pony1: TamePony 2.5, bigAsp 2, Lustify 2

Many thanks to the creators of these wondeful models.

Big Love - XL2.5