๊ฐ€์ž… ํ›„ ์ดˆ๋Œ€ ๋งํฌ๋ฅผ ๊ณต์œ ํ•˜๋ฉด ๋™์˜์ƒ ์žฌ์ƒ ๋ฐ ์ดˆ๋Œ€ ๋ณด์ƒ์„ ๋ฐ›์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.

Suvaditya Mukherjee
@halcyonrayes
๊ฐ€์ž… July 2019
920 ํŒ”๋กœ์ž‰ ์ค‘    2.2K ํŒฌ
i asked claude to implement trellis.2 fully in cuda c++, and it is 5.3x faster than the original pytorch implementation ๐Ÿš€ surprisingly, the glb-export function was the bottleneck, not the neural net itself. moving exporting to the gpu gave more gains than anything else. - zero python runtime, all in c++ โšก๏ธ - cost ๐Ÿ’ต ? ~$7 on claude and ~$60 on compute for testing on h200s through @huggingface jobs - it still gave 2.2x better perf while being bitwise identical ๐ŸŽฏ the code is open-sourced for use here -- use it as a source for your agent to optimize your models, or utilize it directly. happy hacking! ๐Ÿค—
๋” ๋ณด๊ธฐ