llama.cpp
Backend sampling multi output
#25532
Merged

Backend sampling multi output #25532

gaugarg-nv
gaugarg-nv gaugarg-nv requested a review 33 days ago
gaugarg-nv gaugarg-nv requested a review from ggerganov ggerganov 33 days ago
gaugarg-nv gaugarg-nv requested a review from CISC CISC 33 days ago
gaugarg-nv gaugarg-nv requested a review from JohannesGaessler JohannesGaessler 33 days ago
gaugarg-nv gaugarg-nv requested a review 33 days ago
github-actions github-actions added testing
github-actions github-actions added server
github-actions github-actions added ggml
gaugarg-nv gaugarg-nv force pushed from f5e3f22d to 5897112b 29 days ago
ggerganov ggerganov assigned ggerganov ggerganov 29 days ago
ggerganov
ggerganov commented on 2026-07-15
gaugarg-nv gaugarg-nv force pushed from a1b62dd1 to 466696c8 28 days ago
gaugarg-nv
gaugarg-nv gaugarg-nv force pushed from 762a06ec to 89cd8669 9 days ago
otacon6530
danbev
danbev commented on 2026-08-04
gaugarg-nv gaugarg-nv force pushed from 89cd8669 to 4cbb71d3 8 days ago
anvme
Pento95
AmerM137
rankaiyx
danbev
danbev commented on 2026-08-05
ggerganov
ggerganov commented on 2026-08-05
ggerganov
ggerganov commented on 2026-08-05
ggerganov
ggerganov commented on 2026-08-05
gaugarg-nv gaugarg-nv force pushed from 4cbb71d3 to 8d163261 7 days ago
github-actions github-actions added documentation
gaugarg-nv
gaugarg-nv gaugarg-nv force pushed from 8d163261 to 9fcc9117 7 days ago
ggerganov
ggerganov
gaugarg-nv
ggerganov
gaugarg-nv
ggerganov
ggerganov
ggerganov approved these changes on 2026-08-06
danbev
danbev approved these changes on 2026-08-10
gaugarg-nv Enable backend sampling with token speculation
fc4fc9b7
gaugarg-nv Clamp the mask sum before converting it into the sampled index
c186a3fb
gaugarg-nv Add a numeric context parameter declaring the maximum outputs one seq…
62dedf43
gaugarg-nv More fixes
dbcec660
gaugarg-nv Don't reuse memory for output views.
aaa886fe
gaugarg-nv Match dist between CPU and GPU
37d606ec
gaugarg-nv Fix CPU and backend sampling mismatches
4dbed2e6
gaugarg-nv Simpify some of the changes
eecc6571
gaugarg-nv Fix tests on Vulkan
3b112404
gaugarg-nv More test fixes
a5663e1d
gaugarg-nv Rebase changes
9e874789
gaugarg-nv Rebase and address review comments
1826901d
gaugarg-nv Address review comments
d4c699c1
gaugarg-nv Address review comments
9f5d0596
gaugarg-nv gaugarg-nv force pushed from 9fcc9117 to 9f5d0596 2 days ago
ggerganov
ggerganov commented on 2026-08-10
gaugarg-nv Update src/llama-sampler.cpp
0ec96eb5
ggerganov ggerganov merged dd1ea524 into master 2 days ago
gaugarg-nv gaugarg-nv deleted the backend-sampling-multi-output branch 2 days ago

Login to write a write a comment.

Login via GitHub

Assignees
Labels
Milestone