This is an experimental checkpoint that combines some of the best realistic Illustrious models with bigASP 2.5. It worked.
BTW, v2.4 is now available here! Give it a try!
❤️ If you like Snakebite, you can help offset the cost of training:
Why it matters
bigASP has great prompt adherence but it is wildly inconsistent as far as style and composition. It feels like a base model with untapped potential. Dialing in the right settings is like trying to solve a Rubik's Cube.
Illustrious models, on the other hand, lose many concepts when going 3D. They're hit-or-miss even with fairly popular booru tags. But the lighting and composition of these weights are still 👌
I wanted to see if we can get the best of both worlds, and it turns out we kinda can! Careful block merging is the key. We can inject bigASP's output_blocks.0 to acquire much of its conceptual knowledge. Adding its middle_block.2 seems to reduce anatomical issues (otherwise, you'll get a lot of extra arms and fingers.)
First impressions
Best prompt adherence for a realistic SDXL model I've ever seen. And no, not just for smut. (But especially for smut)
Compatible with booru tags as well as natural language. I think a mixed prompt approach is best: I use about 70% tags (no underscores) and 30% natural language in a prompt.
Understands bigASP's style tags to an extent, including stuff like masterpiece quality and 35mm.
It *feels* like something new and worth exploring. Don't sleep on bigASP!
Drawbacks
Illustrious 2.0 models support resolutions up to 1024x1440 or 1024x1536 without horrific stretching of anatomy (e.g. longcat-type torsos), but bigASP's optimal resolution is only 832x1216... and I don't recommend going above that in Snakebite. If you do, the anatomy will be mostly okay (which is surprising), but image composition becomes very odd and unpleasant.
Since we're in a strange new latent space, your existing LoRAs won't work very well. But they're worth retraining.
Recommended Settings
The Turbo variant is better for inference. It's super fast and has slightly improved aesthetics. The non-turbo version is useful for finetuning, and it can produce nice textures if you don't mind waiting 25+ steps.
Turbo
8 or 9 steps
LCM sampler
CFG 1
Custom sigmas below, or simple
Full
20 to 28 steps
Euler ancestral sampler
CFG 3 to 4
Custom sigmas below, or simple
Custom sigma curve (you can use comfyui-kjnodes to apply):
15, 8, 4, 2, 2, 1, 0.4, 0.2, 0
If you're getting mangled limbs, you can often salvage the image by adjusting the first few values of your sigma curve. Here's one that is more stable for certain prompts:
14, 5, 2, 2, 2, 1, 0.4, 0.2, 0If you're still getting body horror, you can try the following quality tags (you'll need ComfyUI-ppm to apply negative weights to your positive prompt):
masterpiece quality, realistic photo, (worst quality,:-1) (mutated,:-1)Snakebite is very responsive to stylistic terms, especially by IL standards. Keep the extra "fluff" to a minimum - almost every token I've tried has a significant impact on the picture.
Finally, I suggest trying the CLIPAttentionMultiply node. If you boost the q and v parameters, it will effectively cause your image to become more "Illustrious-like": cleaner, more stable, but less realistic and (usually) less adherent to the prompt. Set both values to 3 for a very clean image.
BIGASP'S CLIP IS NOW IN PLAY!
In versions 1.3 and up, Snakebite includes a little of bigASP's CLIP, which means you can take advantage of more style prompts. Experiment with different terms to see what works. Personally, I keep it simple - this will usually improve your image without any side effects:
high quality, sharp focusWhich version is for me?
If you're wondering which version of the model to use, here's a TL;DR:
v1.4 = next-level realism, jaw-dropping textures, very stable, slightly less vibrant than previous versions and less capable of non-photographic images
v1.3 = good anatomy, good backgrounds, good coherence
v1.2 = best punchy colors
v1.1 = most influence from bigASP (excluding CLIP), dull colors, a failed experiment TBH
v1.0 = impressively creative but very unstable
If you like the model or use it for further finetuning, please let me know! I'd love to see the results. 💪
Description
Initial release, accelerated version, recommended for inference
FAQ
Comments (5)
First test are looking good ♥. A lot of body horror though. Gotta use more negatives in my next session.
Thanks! Try v1.1 if you haven't already - it's more stable with human anatomy. As for negatives, I recommend using negpip so you can use negative weight with CFG 1 (no loss of speed.) I find adding "(extra limbs:-1)" to the positive prompt works well.
@liftweights thanks will try
I often test realism-focused models like Pony, Illustrious, and SDXL on Civitai. Every day, a ton of merged models get uploaded, but when I test them, I usually can’t figure out what makes them stand out. Most version updates don’t even clarify what’s been improved. I believe the reason behind so many model merges and version updates is just to chase buzz points. There’s often no real improvement, and countless models with no distinct personality get uploaded just for those buzz points. That’s why I place a lot of importance on understanding what sets a merged model apart from others when I test them. If a model doesn’t have something unique or better to offer, there’s no reason to use it. The moment I generated my first image with this model, I could tell right away—it’s unique. Bigasp is a rough and highly realistic model, and it’s tough to get the exact image you want from it. But this model, while rooted in Illustrious, carries Bigasp’s characteristics as well. It’s anatomically more accurate than Bigasp and, despite being an Illustrious model, it’s strikingly realistic. Amazing. There’s a reason to pay attention to this model. Give it a try.
Thank you for the kind words, grrrr! I had the same reaction as you when I first merged these ingredients, not knowing what to expect. It turns out there is enough conceptual overlap between Illustrious and bigASP that they can boost each other - a very pleasant surprise.
Stay tuned for the v1.2 update, which should be out very soon. 🙂