llama.cpp
3686e9d6 - CUDA: only support F32/F16 for GGML_OP_REPEAT (#24533)

Commit

8 days ago

CUDA: only support F32/F16 for GGML_OP_REPEAT (#24533)

References

#24533 - CUDA: only support F32/F16 for GGML_OP_REPEAT

Author

leonardHONG

leonardHONG

Parents

Loading