sycl: support reordered Q4_K/Q5_K/Q6_K MoE MUL_MAT_ID#24452
Merged
Conversation
Extend reordered-weight handling to fused MoE MUL_MAT_ID for Q4_K and Q5_K expert tensors and add Q5_K reordered DMMV coverage. Unsupported 3D reorder cases now fall back instead of aborting.
github-actions
Bot
added
ggml
newjordan
changed the title
Contributor
Author
|
Q6 was not fused in the previous PR, and incorporating it had a trickle down speed increase. |
arthw
approved these changes
Jun 13, 2026
arthw
left a comment
arthw
left a comment
Contributor
There was a problem hiding this comment.
It's good job!
For Qwen3.6-35B-A3B-Q4_K_M.gguf on B60 Ubuntu 24.04:
| Test | fa | Base t/s | Primary t/s | Increase Rate (Primary vs Base) |
|---|---|---|---|---|
| pp512 | 0 | 722.87 | 722.62 | -0.03% |
| pp512 | 1 | 711.10 | 693.31 | -2.50% |
| tg128 | 0 | 62.84 | 68.79 | 9.47% |
| tg128 | 1 | 62.95 | 68.95 | 9.53% |
Thank you!
arthw
added
the
merge ready
papamoose
pushed a commit
to papamoose/llama.cpp
that referenced
this pull request
Jun 27, 2026
* sycl: support reordered Q4_K and Q5_K MoE MUL_MAT_ID Extend reordered-weight handling to fused MoE MUL_MAT_ID for Q4_K and Q5_K expert tensors and add Q5_K reordered DMMV coverage. Unsupported 3D reorder cases now fall back instead of aborting. * sycl: extend MoE reorder to Q6_K mul_mat_id
adrianhoehne
pushed a commit
to adrianhoehne/llama.cpp
that referenced
this pull request
Jul 5, 2026
* sycl: support reordered Q4_K and Q5_K MoE MUL_MAT_ID Extend reordered-weight handling to fused MoE MUL_MAT_ID for Q4_K and Q5_K expert tensors and add Q5_K reordered DMMV coverage. Unsupported 3D reorder cases now fall back instead of aborting. * sycl: extend MoE reorder to Q6_K mul_mat_id
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.
Extends the existing SYCL MoE
mul_mat_idreorder path to Q6_K expert weights.This completes reordered MoE coverage for mixed K-quant MoE models whose down-projection experts are Q6_K. Existing Q4_K/Q5_K behavior is unchanged.
Validation:
test-backend-ops test -o MUL_MAT_ID: 714/714Requirements: