Find what you need for your workflow.

Event Horizon Nexus

Comfyui_Minimax_h3_latent_Upscaler-Plus

Neural latent upscaler for Minimax H3 (24ch). Bypasses costly 5B-param VAE decode/encode. Upscale low-res latents directly, then refine. Accelerates high-res video gen, outperforms naive interp.

ImageFirework — Smoke Exposure (API)

v13 release notes for ImageFirework — Smoke Exposure (API). 180 frames, 24 fps reference, local particle reveal, curved gravity tails. Regression suite 21/21 PASS.

Kming D Style

Works best at 0.95 weight.

Musica2

Musica2 est un modèle d'apprentissage profond conçu pour analyser un clip audio (chanson) et déterminer s'il est dans le rythme (calé) ou hors rythme (décalé). Ce modèle a été entraîné from scratch par Clemylia en...

Vn Lp Y26

Detector YOLO26n train trÃan dataset VN (Roboflow / winter2897). DÃ1ng trong Camera-AI và Space ALPR.

clovisia arknight/克洛维希娅 明日方舟

good work weight: M weight(0.8-1.0),C weight (0.6-1.0)

model.safetensors

[SDXL] FLYX3_MIX (AnimagineXL31)

amazon_sex_low_noise.safetensors

model.safetensors

圣聆初雪 Pramanix the Prerita.safetensors

Lava Style [FLUX+SDXL+1.5]

Trigger (SDXL+1.5): ral-lava Trigger (FLUX): ral-lva

Kuroinu 2 / 黒獣 2 (9 characters)

UkiyoFlow

UkiyoFlow blends the traditional elegance of Ukiyo-e art with a contemporary twist. Perfect for creating vibrant and whimsical images that fuse historical Japanese aesthetics with modern elements.

One obsession

蕾妮 Muu Rennil - Rebellion Gilgamesh.safetensors

illust-xl-pony-convenient-censoring-innertube.safetensors

Renka Style

Works best at 0.75 weight.

model.safetensors