llama.cpp
server : enable packing for d1-2B and d1-Omni-2B models
#32553
Merged

server : enable packing for d1-2B and d1-Omni-2B models #32553

tdakhran
tdakhran tdakhran requested a review 10 hours ago
tdakhran tdakhran requested a review from CISC CISC 10 hours ago
tdakhran tdakhran requested a review 10 hours ago
tdakhran tdakhran requested a review 10 hours ago
tdakhran tdakhran changed the title model : enable packing for d1-2B and d1-2B models model : enable packing for d1-2B and d1-Omni-2B. models 10 hours ago
tdakhran tdakhran changed the title model : enable packing for d1-2B and d1-Omni-2B. models model : enable packing for d1-2B and d1-Omni-2B models 10 hours ago
tdakhran llama-bench : size the batch to the prompt for models without memory
7d4ecc11
tdakhran server : read up to 16 lfm3-d1 questions in one pass
09429a16
tdakhran tdakhran force pushed from 3bbbbcd4 to 09429a16 9 hours ago
github-actions github-actions added documentation
github-actions github-actions added server
tdakhran tdakhran changed the title model : enable packing for d1-2B and d1-Omni-2B models server : enable packing for d1-2B and d1-Omni-2B models 9 hours ago
pwilkin
ggml-gh-bot
tdakhran
ngxson
ngxson commented on 2026-11-18
ngxson
ngxson approved these changes on 2026-11-18
ngxson
ngxson ngxson merged bafa0420 into master 6 hours ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone