Real-ESRGAN (MLX)
Collection
Apple MLX fp16 ports of Real-ESRGAN super-resolution (RRDBNet + SRVGGNetCompact), 5 variants, BSD-3. • 5 items • Updated
Real-ESRGAN-x4plus (MLX)
Apple MLX fp16 port of Real-ESRGAN RealESRGAN_x4plus (RRDBNet, ×4), for super-resolution on Apple Silicon. Converted from the official xinntao/Real-ESRGAN release checkpoint (BSD-3-Clause).
Usage
pip install realesrgan-mlx # https://github.com/xocialize/realesrgan-mlx
realesrgan-mlx -i input.png -o out/ -n RealESRGAN_x4plus
from realesrgan_mlx.pipeline_mlx import make_upsampler, upscale_image
up = make_upsampler("RealESRGAN_x4plus", tile=256) # tile>0 caps memory on large images
out = upscale_image("input.png", up)
make_upsampler downloads these weights automatically.
Details
- Architecture: RRDBNet (
num_feat=64,num_block=23,num_grow_ch=32) - Scale: ×4 · Precision: fp16
- Parity vs PyTorch: full-forward 3.8e-6 (CPU fp32); fp16 vs fp32 golden 1.3e-3.
License
BSD-3-Clause (upstream Real-ESRGAN, Xintao Wang et al.).
- Downloads last month
- 23
Model size
16.7M params
Tensor type
F16
·
Hardware compatibility
Log In to add your hardware
Quantized