llama.cpp
b0311c16 - CUDA: fix padding of GQA to power of 2 in FA (#19115)

Commit

3 days ago

CUDA: fix padding of GQA to power of 2 in FA (#19115)

References

#19115 - CUDA: fix padding of GQA to power of 2 in FA

Author

JohannesGaessler

JohannesGaessler

Parents

Loading