ggml/llama.cpp releases: b10991

TL;DR

The ggml/llama.cpp project released version b10991, which includes the addition of a missing contiguous fast-path and hvx_copy_uu for each run. This update is significant as it enhances performance on specific hardware configurations.

Detailed Summary

The ggml/llama.cpp project released version b10991, which includes the re-addition of a missing contiguous fast-path and hvx_copy_uu for each run. This update affects users on macOS Apple Silicon (arm64) and is part of ongoing improvements to the software. The broader impact benefits developers and users by enhancing performance and functionality in the llamacpp environment.

Key Points

• • hexagon: added back missing contiguous fast-path and hvx_copy_uu for each run
• • Website: <https://llama.app>
• • Attestations: <https://github.com/ggml-org/llama.cpp/attestations/47775478>