Georgi Gerganov 1725e316c1 models : optimize qwen3next graph (#19375) ...

* models : optimizing qwen3next graph

* cont

* wip

* wip

* wip

* wip

* wip

* wip

* wip

* wip

* wip

* wip

* cont : remove redundant q, g chunking

* minor

* minor

* avoid passing masks around

* avoid concats during chunking

* naming + shapes

* update names and use prefix to disable CUDA graphs

2026-02-14 12:57:36 +02:00

..

CMakeLists.txt

docs : Minor cleanups (#19252)

2026-02-02 08:38:55 +02:00

ggml-metal-common.cpp

models : optimize qwen3next graph (#19375)

2026-02-14 12:57:36 +02:00

ggml-metal-common.h

metal : refactor + optimize v2 (#15995)

2025-09-17 20:38:12 +03:00

ggml-metal-context.h

metal : support virtual devices (#18919)

2026-02-02 14:29:44 +02:00

ggml-metal-context.m

metal : fix event synchronization in cpy_tensor_async (#19402)

2026-02-07 07:37:15 +02:00

ggml-metal-device.cpp

metal : update sum_rows kernel to support float4 (#19524)

2026-02-12 11:35:28 +02:00

ggml-metal-device.h

metal : consolidate bin kernels (#19390)

2026-02-07 10:35:56 +02:00

ggml-metal-device.m

metal : fix ACC op (#19427)

2026-02-14 09:54:03 +02:00

ggml-metal-impl.h

metal : update sum_rows kernel to support float4 (#19524)

2026-02-12 11:35:28 +02:00

ggml-metal-ops.cpp

metal : fix ACC op (#19427)

2026-02-14 09:54:03 +02:00

ggml-metal-ops.h

metal : support GGML_OP_SET (#19548)

2026-02-13 07:34:52 +02:00

ggml-metal.cpp

metal : add missing includes (#19348)

2026-02-05 08:05:09 +02:00

ggml-metal.metal

metal : update sum_rows kernel to support float4 (#19524)

2026-02-12 11:35:28 +02:00