llama.cpp
3fe362fe
- gguf-py : use ThreadPoolExecutor when writing tensors
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
272 days ago
gguf-py : use ThreadPoolExecutor when writing tensors - gguf-py : handle (limited) retries for remote tensors
References
compilade/parallel-convert
#12837 - convert : write tensors in parallel
Author
compilade
Parents
d7db1593
Loading