llama.cpp
a0d58553 - cuda : extend GGML_OP_PAD to work with non-cont src0 (#19429)

Commit

25 days ago

cuda : extend GGML_OP_PAD to work with non-cont src0 (#19429) * cuda : extend GGML_OP_PAD to work with non-cont src0 * tests : add permuted pad

References

Author

ggerganov

Parents