llama.cpp
mtp: support for gemma-4 E2B and E4B assistants
#24282

Merged

mtp: support for gemma-4 E2B and E4B assistants #24282

max-krasnyansky merged 4 commits into ggml-org:master from qualcomm:gemma-4-support-smaller-assistants

models: update converter to support smaller assistants

f9102120

models: add masked_embd tensors to gemma4-assist arch

0bc25368

gemma-4: remove temp debug for conversion

be07d144

max-krasnyansky requested a review from

CISC 155 days ago

github-actions added model

github-actions added python

ggerganov approved these changes on 2026-06-08

gemma-4-mtp: filter out masked_embedding tensors during conversion

a9af7cfd

CISC approved these changes on 2026-06-08

max-krasnyansky merged 7d2b45b4 into master 154 days ago

Reviewers

CISC

ggerganov

Assignees

No one assigned

Labels

model python

Milestone

No milestone