llama.cpp 发布: b11163

来源摘要

llama: add llama_batch_ext (#24669) * (wip) add llama_batch_ext * wip * updated design * updated impl * change signature * unused var * demo common_prompt_batch_decode * fix pos * tmp disable test-batch-alloc * fix compat * nits: add const * no more pos_max * add comment about llama_batch_ext_set_embd_state * handle n_embd_out properly * rename api --> embd_token * llama_embd * stub llama_batch_ext_set_embd_state * support both token + embd + state in batch * llama_batch_ext_add_embd * upstream some changes * nits * fix test-batch-alloc * add test for compat **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11163/llama-b11163-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b11163/llama-b11163-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11163/llama-b11163-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11163/llama-b11163-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11163/llama-b11163-bin-ubuntu-arm64.tar.gz) - [Ubuntu s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11163/llama-b11163-bin-ubuntu-s390x.tar.gz) - [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b11163/llama-b11163-bin-ubuntu-vulkan-x64.tar.gz) - [Ubuntu arm64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b11163/llama-b11163-bin-ubuntu-vulkan-arm64.tar.gz) - [Ubuntu x64 (CUDA 12)](https://github.com/ggml-org/llama.cpp/relea