Skip to content
New issue

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

Sign up for GitHub

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

Already on GitHub? Sign in to your account

Jump to bottom

VILA-1.5 details #67

Closed
Lopa07 opened this issue May 17, 2024 · 4 comments
Closed

VILA-1.5 details #67

Lopa07 opened this issue May 17, 2024 · 4 comments

Comments

@Lopa07
Copy link

Lopa07 commented May 17, 2024

  • Where can I found the paper for VILA-1.5?
  • What visual encodes and LMs are used for VILA 1.5 3B, 8B, 13B, and 40B?
@Lopa07 Lopa07 changed the title VILA 1.5 details VILA-1.5 details May 17, 2024
@hkunzhe
Copy link

hkunzhe commented May 20, 2024

You can look at the model configuration files on Hugging Face or the training code in the repository.

@Lopa07
Copy link
Author

Lopa07 commented May 20, 2024

Sorry, I can not find these details. It will be very helpful, if you please post these information here for better visibility.

@yaolug
Copy link
Collaborator

yaolug commented May 20, 2024 •
edited
Loading

You can look at the training scripts under https://github.com/Efficient-Large-Model/VILA/tree/main/scripts/v1_5/release
You can refer to the technical details from the original paper. https://arxiv.org/pdf/2312.07533
We made Section 4.4 the default now.

@Lopa07
Copy link
Author

Lopa07 commented May 20, 2024

Thank you both! This helped.

@Lopa07 Lopa07 closed this as completed May 20, 2024
gheinrich pushed a commit to gheinrich/VILA that referenced this issue Dec 16, 2024
@Efficient-Large-Language-Model
Merge pull request NVlabs#67 from Efficient-Large-Model/dev/fix_shard…
4f906a9 
…ed_eval

Fix bugs in sharded evaluation scripts
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@hkunzhe @Lopa07 @yaolug