vulkan: Update topk_moe fusion to handle gpt's late softmax#16656
Merged
Conversation
github-actions
Bot
added
Vulkan
Contributor
Author
|
CC @am17an I've included the ggml_check_edges change in this PR. |
0cc4m
reviewed
Oct 25, 2025
0cc4m
left a comment
0cc4m
left a comment
Contributor
There was a problem hiding this comment.
I understand what this change is doing, but how do I test it? The topk_moe tests pass before and after this change. Which model architectures correspond to the three modes?
Contributor
Usually I put a debug statement printing the number of nodes fused. We'll need to come up with a better way to assert that the nodes were actually fused |
Contributor
Author
I've added some logging in the latest commit that I use to verify fusion and the effects of graph_optimize. You can see the whole sequence of ops without a sync in between, which implies the fusion is working. Early softmax w/norm: qwen3 |
Contributor
Author
|
I've rebased this and updated it to handle the clamp added in #16655. |
Contributor
|
Are the non-Vulkan changes fine @slaren ? |
slaren
approved these changes
Oct 29, 2025
Co-authored-by: Diego Devesa <slarengh@gmail.com>
Anico2
added a commit
to Anico2/llama.cpp
that referenced
this pull request
Jan 15, 2026
…#16656) * vulkan: Update topk_moe fusion to handle gpt's late softmax Based on ggml-org#16649. * Add ggml_check_edges * Add sync logging to show fusion effects * handle clamp added in ggml-org#16655 * Update ggml/src/ggml-impl.h Co-authored-by: Diego Devesa <slarengh@gmail.com>
blime4
referenced
this pull request
in blime4/llama.cpp
Feb 5, 2026
* vulkan: Update topk_moe fusion to handle gpt's late softmax Based on #16649. * Add ggml_check_edges * Add sync logging to show fusion effects * handle clamp added in #16655 * Update ggml/src/ggml-impl.h Co-authored-by: Diego Devesa <slarengh@gmail.com>
Seunghhon
pushed a commit
to Seunghhon/llama.cpp
that referenced
this pull request
Apr 26, 2026
…#16656) * vulkan: Update topk_moe fusion to handle gpt's late softmax Based on ggml-org#16649. * Add ggml_check_edges * Add sync logging to show fusion effects * handle clamp added in ggml-org#16655 * Update ggml/src/ggml-impl.h Co-authored-by: Diego Devesa <slarengh@gmail.com>
ljubomirj
pushed a commit
to ljubomirj/llama.cpp
that referenced
this pull request
May 6, 2026
…#16656) * vulkan: Update topk_moe fusion to handle gpt's late softmax Based on ggml-org#16649. * Add ggml_check_edges * Add sync logging to show fusion effects * handle clamp added in ggml-org#16655 * Update ggml/src/ggml-impl.h Co-authored-by: Diego Devesa <slarengh@gmail.com>
my-other-github-account
pushed a commit
to my-other-github-account/llama.cpp
that referenced
this pull request
May 15, 2026
…#16656) * vulkan: Update topk_moe fusion to handle gpt's late softmax Based on ggml-org#16649. * Add ggml_check_edges * Add sync logging to show fusion effects * handle clamp added in ggml-org#16655 * Update ggml/src/ggml-impl.h Co-authored-by: Diego Devesa <slarengh@gmail.com>
my-other-github-account
pushed a commit
to my-other-github-account/llama.cpp
that referenced
this pull request
May 15, 2026
…#16656) * vulkan: Update topk_moe fusion to handle gpt's late softmax Based on ggml-org#16649. * Add ggml_check_edges * Add sync logging to show fusion effects * handle clamp added in ggml-org#16655 * Update ggml/src/ggml-impl.h Co-authored-by: Diego Devesa <slarengh@gmail.com>
phibya
pushed a commit
to ziee-ai/llama.cpp
that referenced
this pull request
May 29, 2026
…#16656) * vulkan: Update topk_moe fusion to handle gpt's late softmax Based on ggml-org#16649. * Add ggml_check_edges * Add sync logging to show fusion effects * handle clamp added in ggml-org#16655 * Update ggml/src/ggml-impl.h Co-authored-by: Diego Devesa <slarengh@gmail.com>
fewtarius
pushed a commit
to fewtarius/CachyLLama
that referenced
this pull request
May 30, 2026
…#16656) * vulkan: Update topk_moe fusion to handle gpt's late softmax Based on ggml-org#16649. * Add ggml_check_edges * Add sync logging to show fusion effects * handle clamp added in ggml-org#16655 * Update ggml/src/ggml-impl.h Co-authored-by: Diego Devesa <slarengh@gmail.com>
fukuro-kun
pushed a commit
to fukuro-kun/fukuro-llama-cpp-turboquant
that referenced
this pull request
Jul 5, 2026
…#16656) * vulkan: Update topk_moe fusion to handle gpt's late softmax Based on ggml-org#16649. * Add ggml_check_edges * Add sync logging to show fusion effects * handle clamp added in ggml-org#16655 * Update ggml/src/ggml-impl.h Co-authored-by: Diego Devesa <slarengh@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
4 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.
Based on #16649.