llama.cpp
tools : split Metal FA-vec tuning into a standalone tuner
#26620

Merged

tools : split Metal FA-vec tuning into a standalone tuner #26620

forforever73 merged 10 commits into stack/metal-fa-vec from stack/metal-tuner

forforever73 requested a review from

ggerganov 7 days ago

forforever73 requested a review 7 days ago

add device id -> token lookup for the offline tuning tool

eb052925

add ggml-metal-tuning skeleton

55318e3b

add op-agnostic perf cell + median timing for the tuner

cf80ea68

add FA-vec graph build + tensor init to the tuner

8e7831fd

tools : add FA-vec (Q,NE) sweep, compression and table emit

f44d2ee2

cool down and re-measure the dirty window on thermal drift

35c08cd0

test-backend-ops : replace the FA vec tune mode with a bounded (Q,NE)…

26edf081

tools : document the Metal tuner, point the table comment at it

ff024685

abort on unknown KV type, single-source fa_vec_legal_ne

9be171f5

cleanup

8045779c

forforever73 force-pushed the stack/metal-fa-vec branch from aa4b5996 to 0f279350 7 days ago

forforever73 force pushed from 4c927aee to 8045779c 7 days ago

github-actions added documentation

github-actions added testing

github-actions added examples

github-actions added ggml

github-actions added Apple Metal

ggerganov approved these changes on 2026-08-10

forforever73 merged 8045779c into stack/metal-fa-vec 1 day ago

Reviewers

ggerganov

Assignees

No one assigned

Labels

documentation testing examples ggml Apple Metal

Milestone

No milestone