llama.cpp
55cf48de
- cuda : fix multi-seq, quantized FA
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
189 days ago
cuda : fix multi-seq, quantized FA ggml-ci
References
gg/fix-fa-q-non-cont
Author
ggerganov
Parents
a856a566
Loading
