transformers
22ddccc4
- Attempt to use custom cuda kernels for speed up inference for bloom.
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
3 years ago
Attempt to use custom cuda kernels for speed up inference for bloom.
References
custom_bloom_kernel
Author
Narsil
Parents
8881e58b
Loading