GitLab

Follow up from #12992

Free all streams, and keep the alloc logic aligned across streams.

* Revert "vulkan: temporary cary of vulkan fixes (#12971)"

This reverts commit 3a9e8e9f.

* ggml update to b7087

* fix argsort on metal

* update to b7108

* fix bakllava regression

This model lacks the metadata for the projector type.

* update to b7209

* fix TopK perf

* only build arm code on arm

cuda panics on batches larger than 1024 so skip those and fallback to
cpu