Encuentra lo que necesita tu flujo.

Ada MXAI

Ada is a surprisingly buxom young woman trained on 167 images created solely by prompts and bodysliders, 20 vectors, learning rate 0.004, 20 steps pr image/embedding and 2000 steps in total. Responds well to all...

Qwen3 Asr 0.6b Arabic Dialectal

Fine-tune of Qwen/Qwen3-ASR-0.6B-hf (0.6B (938M)) for multi-dialect Arabic speech recognition (undiacritized output).

flux2-dev_fp8matmulmixed.safetensors

pvorldpthsxnnZIB.safetensors

Sango Style

Works best at 0.75 weight.

Ghost in the Shell 2026

This LyCORIS (eventually) captures the main characters and the general style of the anime.

guruguru / @_@ Concept LoRA

武者巫女 トモエ Tomoe Queen's Blade.safetensors

Krea 2 Turbo Official Comfy-Org Checkpoints (Krea2)

Awesome COMBO

Another model suitable for animations in ComfyUI. It has 7 separate concepts, so it has 7 trigger words :-) You can make crazy stuff with blending concepts in prompts. Check my examples.

model.safetensors

XL_地狱把妹王_Helltaker style

Ddpm Celebahq Finetuned Butterflies 2epochs

This model is a fine-tuned version of DDIM on the butterfly dataset

Stylized PBR Animation [MiniMax H3 Ref2VA]

Creates videos in a stylized physically based rendering animation aesthetic, comparable to the kind seen in major animated films.

TPDD Honglin CLS Yolo26n Cls

tla-derived-noncommercial

model.safetensors

wan22-i2v-dildo-low-104-41.safetensors

ZIT Braces (v1.0)

IQA ONNX Models

Pre-converted ONNX models for No-Reference Image Quality Assessment (NR-IQA), exported from IQA-PyTorch and ready for deployment with ONNX Runtime (no PyTorch dependency needed).

Flux1-DedistilledMixTuned