ggerganov/llama.cpp b7959: b7959
AI Summary
The article summarizes the release of version b7959 of the ggerganov/llama.cpp project, which is a C++ implementation of the LLaMA language model. The key points include: 1. The release includes a new function called `common_speculative_is_compat()` that checks if the target context is compatible for speculative decoding. 2. The release also includes a revert of the `llama_memory_can_rm_suffix()` function, which was previously added. 3. The article lists the various platforms and configurations supported by the release, including macOS (both Apple Silicon and Intel), iOS, Linux (Ubuntu x64, s390x), Windows (x64, arm64, CUDA, Vulkan, SYCL, HIP), and openEuler (x86, aarch64). 4. Overall, this release focuses on improving the compatibility and functionality of the LLaMA language model implementation in C++.
Original Description
<details open> common : add common_speculative_is_compat() (#19270) * llama : add llama_memory_can_rm_suffix() * Revert "llama : add llama_memory_can_rm_suffix()" This reverts commit d30e59b62a15ef4266a6503e3f4eba770aec001b. * spec : check if the target context is compatible for spec decoding </details> **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b7959/llama-b7959-bin-macos-arm64.tar.gz) - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b7959/llama-b7959-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b7959/llama-b7959-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b7959/llama-b7959-bin-ubuntu-x64.tar.gz) - [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b7959/llama-b7959-bin-ubuntu-vulkan-x64.tar.gz) - [Ubuntu s390x (CPU)](https://github.com/ggml-org/llama
Details
Discussion coming soon...