metal: add CONV_3D #19927
Merged
ggerganov merged 4 commits intoggml-org:masterfrom Mar 23, 2026
Merged
Conversation
github-actions
Bot
added
ggml
Ra5hidIslam
force-pushed
the
metal-add-conv3d
branch
from
90df9bc to
3efa70f
Compare
Contributor
Author
|
Hi @ggerganov , just checking in to see if there is anything else needed from my side for this PR. I've rebased it with the latest master to ensure it's ready for review. Thanks! |
ggerganov
approved these changes
Mar 19, 2026
Member
ggerganov
left a comment
ggerganov
left a comment
There was a problem hiding this comment.
I guess a first version of the implementation is OK.
Rebase on latest master and address the comments.
Ra5hidIslam
force-pushed
the
metal-add-conv3d
branch
from
3efa70f to
e0b47d9
Compare
Ra5hidIslam
requested review from
a team,
CISC,
IMbackK,
ngxson and
pwilkin
as code owners
github-actions
Bot
added
documentation
Member
|
@Ra5hidIslam Bad rebase. :) |
CISC
removed
documentation
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Rebased with master and resolved conflicts.
Ra5hidIslam
force-pushed
the
metal-add-conv3d
branch
from
1da7055 to
34358e4
Compare
…ed back, should pass all tests now
Contributor
Author
|
@ggerganov Hi, the ubuntu-cpu (s390x) job seems to have failed due to the self-hosted runner losing connection — doesn't appear to be a code issue. Would a re-run of the failed jobs fix this? |
Seunghhon
pushed a commit
to Seunghhon/llama.cpp
that referenced
this pull request
Apr 26, 2026
* Apply suggestions from code review Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * metal:add conv_3d backend Rebased with master and resolved conflicts. * Resolved issues related to changes in variable names * kernel void kernel_upscale_bilinear_f32 was missing in my branch, added back, should pass all tests now --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
rsenthilkumar6
pushed a commit
to rsenthilkumar6/llama.cpp
that referenced
this pull request
May 1, 2026
* Apply suggestions from code review Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * metal:add conv_3d backend Rebased with master and resolved conflicts. * Resolved issues related to changes in variable names * kernel void kernel_upscale_bilinear_f32 was missing in my branch, added back, should pass all tests now --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds F32 and F16 weight support for the GGML_OP_CONV_3D operation in the Metal backend.
The implementation matches CPU semantics for 3D spatial convolutions (handling strides, padding, and dilation) and enables both weight types to pass functional tests.
Passed Tests
Performance tests don't exits.
Part of #14909