Jeff Bolz 0a99b4c377 vulkan: handle quantize_q8_1 overflowing the max workgroup count (llama/18515) ...

* vulkan: handle quantize_q8_1 overflowing the max workgroup count

* vulkan: Fix small tile size matmul on lavapipe

* fix mul_mat_id failures

2026-01-14 09:11:59 +02:00

..

ggml-blas

…

ggml-cann

CANN: add operator fusion support for ADD + RMS_NORM (llama/17512)

2026-01-14 09:11:59 +02:00

ggml-cpu

kleidiai: add and integrate SVE 256-bit vector-length kernel (llama/18458)

2025-12-31 17:52:09 +02:00

ggml-cuda

sampling : add support for backend sampling (llama/17004)

2026-01-14 09:11:59 +02:00

ggml-hexagon

ggml-hexagon: optimize activation function (llama/18393)

2026-01-14 09:11:59 +02:00

ggml-hip

…

ggml-metal

metal : adjust extra size for FA buffer to avoid reallocations (llama/18545)

2026-01-14 09:11:59 +02:00

ggml-musa

…

ggml-opencl

opencl: allow resizing transpose buffers (llama/18384)

2025-12-31 17:52:09 +02:00

ggml-rpc

rpc : use unordered_map::reserve and emplace (llama/18513)

2026-01-14 09:11:59 +02:00

ggml-sycl

sycl: add newline at the end of CMakeLists.txt (llama/18503)

2025-12-31 17:52:09 +02:00

ggml-vulkan

vulkan: handle quantize_q8_1 overflowing the max workgroup count (llama/18515)

2026-01-14 09:11:59 +02:00

ggml-webgpu

ggml webgpu: unary op suppport, code refactoring, ops support (llama/17764)

2025-12-12 17:53:18 +02:00

ggml-zdnn

…

ggml-zendnn

ggml-zendnn : add ZenDNN backend for AMD CPUs (llama/17690)

2025-12-12 17:53:21 +02:00

CMakeLists.txt

kleidiai: add and integrate SVE 256-bit vector-length kernel (llama/18458)

2025-12-31 17:52:09 +02:00

ggml-alloc.c

llama: automatically set parameters not set by the user in such a way that maximizes GPU utilization (llama/16653)

2025-12-18 08:20:56 +02:00

ggml-backend-impl.h

…

ggml-backend-reg.cpp

ggml-zendnn : add ZenDNN backend for AMD CPUs (llama/17690)

2025-12-12 17:53:21 +02:00

ggml-backend.cpp

vulkan: extend topk_moe to handle sigmoid w/exp_probs_b for nemotron (llama/18295)

2026-01-14 09:11:59 +02:00

ggml-common.h

…

ggml-impl.h

cmake: Added more x86_64 CPU backends when building with GGML_CPU_ALL_VARIANTS=On (llama/18186)

2025-12-31 17:52:09 +02:00

ggml-opt.cpp

…

ggml-quants.c

…

ggml-quants.h

…

ggml-threading.cpp

…

ggml-threading.h

…

ggml.c

llama: automatically set parameters not set by the user in such a way that maximizes GPU utilization (llama/16653)

2025-12-18 08:20:56 +02:00

ggml.cpp

…

gguf.cpp

…