llama.cpp
spec: enable backend sampling for both dflash & dspark
#26958

Merged

spec: enable backend sampling for both dflash & dspark #26958

ggerganov merged 3 commits into ggml-org:master from ruixiang63:dflash_gpu_sampling

dflash: enable backend sampling for both dflash & dspark

38d063c0

ruixiang63 requested a review 5 days ago

ruixiang63 removed review request 5 days ago

ruixiang63 requested a review from

gaugarg-nv 5 days ago

ruixiang63 requested a review from

ggerganov 5 days ago

enable p_min > 0 in backend sampling and add guard

4afbaac2

pcuenca commented on 2026-08-13

ggerganov commented on 2026-08-13

ggerganov approved these changes on 2026-08-13

pcuenca approved these changes on 2026-08-13

cont : add TODO

8ef4fe12

ggerganov merged 0d0bfcd4 into master 4 days ago

ruixiang63 deleted the dflash_gpu_sampling branch 4 days ago

Reviewers

ggerganov

pcuenca

gaugarg-nv

Assignees

No one assigned

Labels

None yet

Milestone

No milestone