]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 06961e2)
sampling : remove sampling branching in output_reserve (#18811)
authorDaniel Bevenius <redacted>
Wed, 28 Jan 2026 04:59:30 +0000 (05:59 +0100)
committerGitHub <redacted>
Wed, 28 Jan 2026 04:59:30 +0000 (05:59 +0100)
commiteef375ce167cf36d8ed2cf347be95a7a7a90fb5e
tree31e33ec0fcf6a5fc67cb3829c3b31fd69688071ctree
parent06961e2876c7ed5ebe43c7e7c9b284dedba567c9commit | diff
sampling : remove sampling branching in output_reserve (#18811)

* sampling : remove sampling branching in output_reserve

This commit updates output_reserve in llama-context.cpp to always
allocate sampling buffers regardless of whether sampling is needed for
the current batch.

The motivation for this is to avoid reallocations and branching based on
the sampling requirements of the batch.
src/llama-context.cpp diff | blob | history
src/llama-context.h diff | blob | history
Packaging of ggml-org/llama.cpp