Skip to content

sync: minja #11352

New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

Merged
merged 1 commit into from
Jan 22, 2025
Merged

sync: minja #11352

merged 1 commit into from
Jan 22, 2025

Conversation

ochafik
Copy link
Collaborator

@ochafik ochafik commented Jan 22, 2025

Template of MiniMaxAI/MiniMax-Text-01 fixed (in --jinja mode) by google/minja#29 (ref, cc/ @fairydreaming )

@ochafik ochafik changed the title contrib: sync minja to support MiniMaxAI/MiniMax-Text-01 template w/ --jinja sync: minja Jan 22, 2025
@ochafik ochafik marked this pull request as ready for review January 22, 2025 15:33
@fairydreaming
Copy link
Collaborator

I tested combination of #11016 and #11352 in my minimax-text-01 branch and:

  • llama.cpp no longer crashes during model load
  • after passing -p "You are a helpful assistant" -cnv --jinja to llama-cli I can chat with the model (without --jinja it throws exception "this custom template is not supported" as expected)

jinja support is pretty cool, thanks for implementing this!

@ochafik ochafik merged commit c64d2be into ggml-org:master Jan 22, 2025
45 checks passed
anagri pushed a commit to BodhiSearch/llama.cpp that referenced this pull request Jan 26, 2025
tinglou pushed a commit to tinglou/llama.cpp that referenced this pull request Feb 13, 2025
arthw pushed a commit to arthw/llama.cpp that referenced this pull request Feb 26, 2025
mglambda pushed a commit to mglambda/llama.cpp that referenced this pull request Mar 8, 2025
# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants