Local Ai
b8738
llama.cpp release **b8738** is a build from the [ggml-org/llama.cpp](https://github.com/ggml-org/llama.cpp) project introducing experimental backend-agnostic tensor parallelism, enabled via the `--...
llama.cpp release b8738 is a build from the ggml-org/llama.cpp project introducing experimental backend-agnostic tensor parallelism, enabled via the --split-mode tensor flag (PR #19378). In practice, only the CUDA backend has received the necessary extensions and performance optimizations to make this mode outperform --split-mode layer in some scenarios. The feature is explicitly marked experimental and not yet production-ready.
Related
Source: local-ai