]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 5788b51)
speculative : refactor enabled configs common_speculative_init (#26510)
authorDaniel Bevenius <redacted>
Tue, 4 Aug 2026 11:17:15 +0000 (13:17 +0200)
committerGitHub <redacted>
Tue, 4 Aug 2026 11:17:15 +0000 (13:17 +0200)
commit7bd8282c37fcd9c4d7236106d664761a23318f18
tree193cdb81827abab6c4454cbac95917996e9456f1tree
parent5788b510a1e3394fcc2d6b13ba2d9d8fc4a5c139commit | diff
speculative : refactor enabled configs common_speculative_init (#26510)

This commit contains a suggestion to reduce some code duplication in
common_speculative_init when adding the enabled speculative decoding
configurations.

No tests were added but the existing server tests still passes with this
change:
```console
$ ./tests.sh unit/test_speculative.py -v -x
```
common/speculative.cpp diff | blob | history
Packaging of ggml-org/llama.cpp