All models output 4× resolution. For 2× output, the upscaler runs at 4× then applies high-quality Lanczos downscaling.
Usage in FFMPEGA
Set
llm_model
→
none
Set
no_llm_mode
→
ai_upscale
Choose
upscale_model
(e.g.
hat_x4
for best quality)
Choose
upscale_scale
(
4
or
2
)
Connect an image or video input and run
Model Loading
Models are loaded via
spandrel
, which auto-detects the architecture from the checkpoint file. No additional dependencies are needed beyond what ComfyUI already provides.
ai-upscale-models huggingface.co is an AI model on huggingface.co that provides ai-upscale-models's model effect (), which can be used instantly with this AEmotionStudio ai-upscale-models model. huggingface.co supports a free trial of the ai-upscale-models model, and also provides paid use of the ai-upscale-models. Support call ai-upscale-models model through api, including Node.js, Python, http.
ai-upscale-models huggingface.co is an online trial and call api platform, which integrates ai-upscale-models's modeling effects, including api services, and provides a free online trial of ai-upscale-models, you can try ai-upscale-models online for free by clicking the link below.
AEmotionStudio ai-upscale-models online free url in huggingface.co:
ai-upscale-models is an open source model from GitHub that offers a free installation service, and any user can find ai-upscale-models on GitHub to install. At the same time, huggingface.co provides the effect of ai-upscale-models install, users can directly use ai-upscale-models installed effect in huggingface.co for debugging and trial. It also supports api for free installation.