llama.cpp
Include speculative decoding stats when timings_per_token is enabled
#12603

Merged

Include speculative decoding stats when timings_per_token is enabled #12603

ggerganov merged 3 commits into ggml-org:master from mostlygeek:master

Include speculative decoding stats when timings_per_token is true

2dc29181

mostlygeek requested a review from

ngxson 1 year ago

github-actions added examples

github-actions added server

ngxson approved these changes on 2025-03-27

ggerganov commented on 2025-03-27

Remove redundant draft_accept_ratio var

41a8e85d

add draft acceptance rate to server console output

429820ec

mostlygeek closed this 1 year ago

mostlygeek reopened this 1 year ago

ggerganov approved these changes on 2025-03-28

mostlygeek requested a review from

ggerganov 1 year ago

ggerganov merged 5d016702 into master 1 year ago

Reviewers

ggerganov

ngxson

Assignees

No one assigned

Labels

examples server

Milestone

No milestone