zephyr

Add fast high resolution marching cubes up to 1024^3.

18

Changes: * Do triplane sampling in a block based manner, reduced VRAM usage. This version can generate 1024^3 on 3090 in 33 seconds, use only 13GB VRAM. Actually this version...

Jagged edges when using high resolution marching cubes.

2

I set marching cubes resolution to highest(320), there is significant jagged edges.  The model outputs density fields, I think using distance field will get better mesh quality?

Do you plan making the rpsl_hlslc open source?

4

I think the compiler need to be customizable.

[Bug]: Using SillyTavern slow down aphrodite generation speed.

3

### Your current environment ```text PyTorch version: 2.3.0+cu121 Is debug build: False CUDA used to build PyTorch: 12.1 ROCM used to build PyTorch: N/A OS: Ubuntu 22.04.3 LTS (x86_64) GCC...