PR #25173 spec: add DSpark speculative decoding

spec: add DSpark speculative decoding #25173

wjinxu wants to merge 5 commits into ggml-org:master from wjinxu:dspark-upstream

github-actions added model

github-actions added conversion

wjinxu force pushed from f3b83cdf to d74ff774 8 days ago

github-actions added testing

wjinxu force pushed from d74ff774 to 37f2513d 8 days ago

wjinxu marked this pull request as ready for review 7 days ago

wjinxu requested a review from

JohannesGaessler 7 days ago

wjinxu requested a review from

CISC 7 days ago

wjinxu requested a review from

ggerganov 7 days ago

wjinxu requested a review 7 days ago

CISC requested a review from

ruixiang63 7 days ago

CISC commented on 2026-06-30

wjinxu force pushed from 37f2513d to aae2941a 7 days ago

wjinxu force pushed from aae2941a to d8b38f23 7 days ago

lym000000 commented on 2026-07-01

wjinxu force pushed from d8b38f23 to 47f34420 7 days ago

wjinxu marked this pull request as draft 6 days ago

spec: add DSpark speculative decoding

96c5be9d

wjinxu force pushed from 47f34420 to 96c5be9d 6 days ago

wjinxu marked this pull request as ready for review 6 days ago

am17an commented on 2026-07-04

spec: read draft block size in the dflash impl

8c548e7a

docs: add DSpark section to speculative.md

47932db0

spec: keep dspark block size read in the dspark impl

1ba891a0

github-actions added documentation

am17an approved these changes on 2026-07-08

dspark : add TODOs for incomplete parts

18290202

ggerganov assigned

ggerganov 7 hours ago

ggerganov unassigned

ggerganov 24 minutes ago

ruixiang63 commented on 2026-07-08

Reviewers

am17an

ruixiang63

CISC

lym000000

JohannesGaessler

ggerganov

Assignees

No one assigned

Labels

documentation model testing conversion

Milestone

No milestone

llama.cpp spec: add DSpark speculative decoding #25173 Open

spec: add DSpark speculative decoding #25173

llama.cpp
spec: add DSpark speculative decoding
#25173

Open