Skip to content

convert : fix Gemma3N, GraniteMoe and Ernie4.5Moe#19084

Merged
CISC merged 2 commits intomasterfrom
cisc/convert-gemma3n-yield-fix
Jan 25, 2026
Merged

convert : fix Gemma3N, GraniteMoe and Ernie4.5Moe#19084
CISC merged 2 commits intomasterfrom
cisc/convert-gemma3n-yield-fix

Conversation

@CISC
Copy link
Copy Markdown
Member

@CISC CISC commented Jan 25, 2026

cont #19064

Fixes #19079

@CISC CISC requested a review from am17an January 25, 2026 08:41
@CISC CISC changed the title convert : fix Gemma3N and Ernie4.5Moe convert : fix Gemma3N, GraniteMoe and Ernie4.5Moe Jan 25, 2026
@CISC CISC merged commit 70d8608 into master Jan 25, 2026
9 checks passed
@CISC CISC deleted the cisc/convert-gemma3n-yield-fix branch January 25, 2026 12:05
shaofeiqi pushed a commit to qualcomm/llama.cpp that referenced this pull request Feb 6, 2026
* fix Gemma3N and Ernie4.5Moe

* fix GraniteMoe
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

python python script changes

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Eval bug: convert_hf_to_gguf.py: ValueError when converting Gemma3n models due to double tensor mapping

2 participants