Wan2.2 14B Fast
app_lora is base
app_lora is base
Fast, multi-speaker TTS (44.1kHz) with voice cloning
Chat with multimodal gemma-3-12b-it or gemma-3-4b-it models
Generate high-quality images from text prompts
app is basic torchao quant
app_lora is base
Generate images using prompts and LoRA models
wan2_fast.py is wan22, app.py is wan2.1 with light lora
For z image
app_quant_latent.py is now main
Generate images from latent tensors
Generate code or answers from text prompts
Generate text responses based on user prompts
generate a video from a text prompt
generate a video from a text prompt
Generate Hindi speech from text using specific speakers
use app quick to gene in .1 sec app.py is ,main
Generate Hindi speech with emotional tones
app_kv takes .3 GPU sec app_kv_1.py is Fa3 with 8bit
GPT OSS docker to Gra
Create powerful AI models without code
Run attention kernel tests interactively
app.py is normal cpu other is gpu