llama.cpp
Adreno optimization for MoE - MxFP4
#22301
Merged

Adreno optimization for MoE - MxFP4 #22301

shawngu-quic
shawngu-quic shawngu-quic requested a review from CISC CISC 96 days ago
shawngu-quic shawngu-quic requested a review 96 days ago
ggml-gh-bot
github-actions github-actions added ggml
github-actions github-actions added OpenCL
CISC
CISC commented on 2026-04-24
shawngu-quic MoE Mxfp4 CLC kernel added, router reorder on GPU
1e1e46e1
shawngu-quic Pass test-backend-ops for MoE mxfp4 Adreno CLC
af39e4d7
shawngu-quic remove putenv in llama-model.cpp
2120deba
shawngu-quic fix indent style and whitespace
a210d8f4
lhez opencl: remove unnecessary headers
5dcfd7cf
lhez opencl: do not save cl_program objects
fea8350c
lhez opencl: remove unnecessary assert
ae9d5eb8
shawngu-quic fix precision issue
9cc5e90a
lhez lhez force pushed from 2a099706 to 9cc5e90a 89 days ago
lhez
lhez approved these changes on 2026-05-01
lhez
CISC
CISC approved these changes on 2026-05-01
lhez lhez merged c5a3bc39 into master 87 days ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone