⚡ EARLY ACCESS — Limited time only. This model is available for direct download for 15 days only. After that: Generation Only on Civitai. No more downloads here. Get the best pricing in the official store.
→ Get it now for much cheaper prices, skip the wait:
🛒 Ko-fi · TensorArt ✅ Card & PayPal accepted — no crypto, no Buzz, no queues ✅ Instant download · Priority support
Also available:
Event Horizon Nexus XL Aenima v1.0 ( Kofi)
Event Horizon Nexus ZIT 2.0 (Kofi)
💎 Get the complete Event Horizon collection with all the future updates included→ Ko-fi
⚙️ Recommended settings with Z-Image Turbo versions:
8 steps, cfg1, Euler or DPM++ 2s a RF sampler, Normal or Simple or Beta scheduler, 832x1216, 1024x1536 or 1280x1920 pixels
Upscaling:
8 steps, cfg1, Euler sampler, Normal or Simple or Beta scheduler, 1.5x upscale
Important: remember that you need to place qwen3_4b.safetensors in the text_encoder sub folder and ae.safetensors in the VAE sub folder for Z-Image Turbo to work properly.
⚙️ Recommended settings with SDXL AENIMA version
Sampling method: DPM++ 2M SDE Karras
VAE: is already Baked In
Sampling steps: 30+ Steps
Resolution: 1344x1792 / 1408x1888 CFG: 3-5
Clip Skip: 2
Upscale: 1.5 upscale (0.3 - 0.35 denoise)
Upscalers: 4x_NickelbackFS_72000_G / 4x_foolhardy_Remacri / 4x_NMKD-Siax_200k
Dmd2 lora: (dmd2 lora is highly recomended with SDXL) LCM Karras scheduler, 14 steps, CFG:1)
⚙️ Recommended settings with SDXL version
Sampling method: DPM++ 2M SDE Karras
VAE: is already Baked In
Sampling steps: 30+ Steps
Resolution: 832x1216 / 896x1152 CFG: 3-5
Clip Skip: 2
Upscale: 1.5 upscale (0.3 - 0.35 denoise)
Upscalers: 4x_NickelbackFS_72000_G / 4x_foolhardy_Remacri / 4x_NMKD-Siax_200k
⚙️ Recommended settings with DMD2 lora:
Sampler: LCM Karras / Exponential
CFG: 1-1.3 (recommended is 1)
Steps: 8 - 12 steps
Resolution: 896x1152 / 832x1216
Clip Skip: 2
VAE is bundled in, you don't need to download anything else.
Description
Better skin detail. No more "airbrushed" skin.
Much better anatomy. Decreased number of third arms / third hands appearing out of nowhere vs 1.0.
Significantly better at nsfw.
FAQ
Comments (11)
No bf16 version for ZIT?
No, sorry. I always release zit checkpoints on fp8.
@ElectricDreams I can tell based on your generations that a 16 version and some tweaks to prompts and user settings would make your photos look even better and even more realistic. I hope you change your mind.
Man, I bought your model on whoop, because I thought that would be in bf16. Now my money is gone and I’m mad, because it’s fp8 only. I don’t want this quantised bs. -.-
@CupRunethOverWithLSD Hi,
I understand your frustration. However, all my Event Horizon models (including this one) are FP8. This is clearly mentioned on Civitai in the preview version, and it's the format I use across all my releases because it offers the best balance between quality, speed and file size.
In my tests and from user feedback, the FP8 version actually performs better in real use than many larger BF16 models due to the specific optimizations I apply. It's not a "quantized downgrade" — it's a deliberate choice for performance.
I'm sorry if the format wasn't 100% clear on the Whop page (I'll improve that). Unfortunately, as this is a digital download, I can't offer a refund once the file has been downloaded.
If you want help getting the best results with this model or have any other questions, I'm here to assist.
@ElectricDreams It's a shame because you can't do anything with a quantized version at all. I - for example - like to merge my perfect models together myself, and this takes that option away—which is really a shame.
@CupRunethOverWithLSD Hi,
I understand your concern about merging. That's a valid point for some users.
All my Event Horizon ZIT models are intentionally released in optimized FP8. This is not a heavy quantization that destroys quality — it's a deliberate choice that gives excellent visual results while being much faster and lighter. Many people actually prefer FP8 for daily use and merging because of the speed.
That said, you're right that merging FP8 with BF16 models is not ideal and can cause issues. FP8 works best when merged with other FP8 models.
If merging is your main workflow, my models might not be the best fit for you right now. I focus on delivering strong, ready-to-use checkpoints rather than base models for heavy merging.
I don't currently offer BF16 versions because they would be roughly double the size (10-13GB each), and with my current storage situation it's not sustainable. I may offer them in the future once I upgrade my infrastructure.
Sorry if this doesn't match what you were looking for. If you have any other questions about using the model, feel free to ask.
@ElectricDreams I would appreciate a change, 10GB is a tiny size after all. If you want to sell your checkpoints you probably sell more if you offer bf16. Look at other checkpoints and the bf16 version usually have about 4x more downloads then the fp8 version.
@Whistler42 Dully noted. Thanks for your feedback. If the opportunity to greatly expand my current store infraestructure appears it could be an option. I have to mull over it.
what's the difference between nexus and the regular one for ZIT?
Hi, yes, EH ZIT focus on artistic styles and EH ZIT NEXUS focus on better anatomy and nsfw.