Sounds like these guys didn't use custom kernels, but BitNet did. | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		thatguysaguy on March 29, 2024 \| parent \| context \| favorite \| on: Towards 1-bit Machine Learning Models Sounds like these guys didn't use custom kernels, but BitNet did.

mobicham on March 30, 2024 [–]

That's correct. Only the dequantization is done on CUDA, the matmul is done with Pytorch. If they put their kernels open-source we could re-use them!

Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10
Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact