]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: fb78ad2)
model : fix Granite Hybrid type check for 7B.A1B (#20795)
authorVictor Villar <redacted>
Fri, 20 Mar 2026 14:16:09 +0000 (15:16 +0100)
committerGitHub <redacted>
Fri, 20 Mar 2026 14:16:09 +0000 (15:16 +0100)
commit58c81f7e811d82b9ba62002f40d7af56134b61fc
tree9a47382c5331cc8cac2c57251cbb841385ab0092tree
parentfb78ad29bbe7ae00619b2ce31b0a71e95fdbfc43commit | diff
model : fix Granite Hybrid type check for 7B.A1B (#20795)

* Check granite hybriid expert count to set type as LLM_TYPE_7B_A1B or LLM_TYPE_1B

* Use feed fwd dim instead of num of experts

Co-authored-by: Sigbjørn Skjæret <redacted>
---------

Co-authored-by: Sigbjørn Skjæret <redacted>
src/llama-model.cpp diff | blob | history
Packaging of ggml-org/llama.cpp