llama.cpp
ggml : add BF16 support to RMS_NORM (CPU, CUDA)
#32135

Merged

ggml : add BF16 support to RMS_NORM (CPU, CUDA) #32135

ggerganov merged 2 commits into ggml-org:master from anujj:anujj/bf16-rms-norm

ggml : add BF16 support to RMS_NORM (CPU, CUDA)

576199f5

anujj marked this pull request as ready for review 3 days ago

anujj requested a review from

ggerganov 3 days ago

anujj requested a review 3 days ago

anujj requested a review from

wine99 3 days ago

anujj requested a review 3 days ago

am17an commented on 2026-11-12

am17an approved these changes on 2026-11-12

github-actions added testing

github-actions added Vulkan

github-actions added ggml

github-actions added SYCL

github-actions added Apple Metal

github-actions added Ascend NPU

github-actions added OpenCL

github-actions added CUDA

github-actions added OpenVINO

cuda : use one helper for the F32/BF16 RMS_NORM launches

7263b767

ggerganov approved these changes on 2026-11-13

ggerganov merged 31953bcf into master 2 days ago

Reviewers

ggerganov

am17an

wine99

Assignees

No one assigned

Labels

testing Vulkan ggml SYCL Apple Metal Ascend NPU OpenCL CUDA OpenVINO

Milestone

No milestone