
Forha
@Forhanvv • 2,029 subscribers
Exploring overlooked Web3 | Wallet drains, protocol risks & privacy tech | FHE enthusiast | AI & promo video creator
Shorts
Videos

THIS 150MB AI CAN SWAP CHARACTERS IN VIDEOS WITH ONE PHOTO 🤯 akatz labs made a MiniMax H3 character-swap LoRA • video and reference photo • keeps the original scene • works with ComfyUI • only 148MB • free Hugging Face demo HOW TO USE: 1. upload your video 2. upload a character photo 3. describe the character you want to replace 4. generate try demo: model: Credit: peng
Forha51,705 Aufrufe • vor 6 Tagen
0:28
Sensitive content
This media may contain sensitive content.

FOUND ANOTHER UNCENSORED LOCAL VIDEO MODEL 💀 This one comes with no restrictions for generating adult content. LTX 2.3 Uncensored: • 22B parameters • Supports text-to-video • Supports image-to-video • Runs locally through ComfyUI • Q4_K_M version is around 14GB HOW TO USE IT: Model: 1. Install ComfyUI 2. Install the required GGUF loader 3. Download the model and all required files 4. Import the included workflow 5. Enter your prompt and start generating Public examples are available in the video.
Forha48,271 Aufrufe • vor 7 Tagen

FOUND AN UNCENSORED LOCAL ANIME IMAGE MODEL 👀 Noct Q Anime is a Qwen Image 2.1-based model built specifically for uncensored anime generation. • anime text-to-image • adult + NSFW generation • detailed prompts + complex scenes • supports multiple characters + poses • 7.3GB INT8 checkpoint • runs locally with ComfyUI • ready-made workflow included RECOMMENDED SETUP: → ComfyUI 0.37+ → 8–12GB VRAM → 25 steps → Euler / Simple → CFG 3 It also includes the workflow and required model components to get started. IMPORTANT: the author says it's non-commercial only under the Qwen Research License. Check it out on Hugging Face Test it locally with fictional adult characters.
Forha20,933 Aufrufe • vor 5 Tagen
Keine weiteren Inhalte verfügbar