⚡ 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 ZIT 2.0 (Kofi)
Event Horizon XL 7.0 (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 versions 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.
⚙️ Normal SDXL Recommended Settings:
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
Description
V 3.0 is a non DMD checkpoint although using the DMD2 lora is recommended to get the most IQ. It should be trainable for loras now.
More photorealistic than 2.5 but without sacrificing creativity. Looking forward to your feedback.
FAQ
Comments (13)
Beautiful, I liked it.
This checkpoint is good. However, I bought your NSFW checkpoint in Tensorart and the results are very burned, no matter what method (DMD2 or not, with or without loras). Are you sure you linked the correct version for the download?
Hi Sanchez! I've just tested it online on TR and the image is alright (it's the same as the download). Are you using other loras with it? if so, if the loras suffer from overfitting can burn the image. I'd recommend to lower the weight of the lora. Try to keep cfg to 1 or very low. LCM Karras, 12 steps. No need to use dmd2 lora with it either. If you use it over the checkpoint it will burn the image. Hope it helps!
Edit: I just downloaded to be doubly sure and it works perfectly fine if you follow the recommended settings.
@ElectricDreams Ah ok. Yes, I've had the DMD2 lora active with it. I'll try without it again!
@Sanchez3 yeah that's what i thought. Enjoy and thanks for your support!
Those are my settings and I am getting blurred images, any help please?
https://ibb.co/0RwcCQbd
You need to use the dmd2 lora with lcm karras, this is not a dmd2 checkpoint, that's why you're having blurry outputs. You can use other sampler methods if you want.
@ElectricDreams How can I do that in SwarmUI?
@VigoVonHomburg Sorry i just don't use swarmUI. But use normal settings like this
Sampling method: DPM++ 2M SDE Karras
VAE: is already Baked In
Sampling steps: 30+ Steps
Resolution: 832x1216 / 896x1152 CFG: 3-5
Clip Skip: 2
It will work too.
@ElectricDreams I got the dmd2 lora, but it seems the generations are "too bright"? I am using the settings mentioned before
@VigoVonHomburg then lower the cfg to 1. Try lower steps. Check out if you're not loading DMD2 lora twice. Hope it helps!
@ElectricDreams It helped, thanks! One more question, is there a way to force a cetrain style? For example a painting, drawing, anime, manga etc? Or do I need to use LORAs?
@VigoVonHomburg no, you can throw any artistic style into Event Horizon 3.0. it will do just fine. I suggest you check out my embeddings for EH 3.0, though. Lots of artistic tools. Have fun!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.