llama.cpp
cuda: GET_ROWS quants
#25962
Merged

cuda: GET_ROWS quants #25962

ServeurpersoCom
ServeurpersoCom cuda: add k-quant support to GET_ROWS
a29b346f
ServeurpersoCom cuda: add i-quant support to GET_ROWS
2e5ff4fe
ServeurpersoCom cuda: add mxfp4 support to GET_ROWS
c2c65892
ServeurpersoCom ServeurpersoCom requested a review 75 days ago
ServeurpersoCom ServeurpersoCom requested a review from am17an am17an 75 days ago
ServeurpersoCom ServeurpersoCom requested a review from pwilkin pwilkin 75 days ago
github-actions github-actions added ggml
github-actions github-actions added CUDA
pwilkin
pwilkin requested changes on 2026-07-21
ServeurpersoCom cuda: gate the GET_ROWS row size only for 32-value sub-block types
fb9cff5e
ServeurpersoCom ServeurpersoCom requested a review from pwilkin pwilkin 75 days ago
pwilkin
pwilkin approved these changes on 2026-07-21
am17an
am17an approved these changes on 2026-07-22
ServeurpersoCom ServeurpersoCom merged c5a4a0bb into master 75 days ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone