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

[FEATURE] Support Internlm-XComposer2-VL Model #590

Open
thiner opened this issue Mar 13, 2024 · 1 comment · May be fixed by #619
Open

[FEATURE] Support Internlm-XComposer2-VL Model #590

thiner opened this issue Mar 13, 2024 · 1 comment · May be fixed by #619
Labels
enhancement New feature or request

Comments

@thiner
Copy link

thiner commented Mar 13, 2024

Is your feature request related to a problem? Please describe.
No.

Describe the solution you'd like
Please support internlm/internlm-xcomposer2-vl-7b-4bit model. They have already provided 4-bit quantized model, and there is sample code for running the model with AutoGPTQ: https://github.com/InternLM/InternLM-XComposer/tree/main#4-bit-model

Describe alternatives you've considered
no.

Additional context
Add any other context or screenshots about the feature request here.
@thiner thiner added the enhancement New feature or request label Mar 13, 2024
@thiner thiner changed the title [FEATURE] Support Internlm- [FEATURE] Support Internlm-XComposer2-VL Model Mar 13, 2024
@LaaZa
Copy link
Contributor

LaaZa commented Mar 27, 2024 •
edited

This is a bit of a problem, since internlm modeltype is already supported in AutoGPTQ but they are using completely different layout to normal InternLM. It would be possible to use a custom modeltype name but that would require changing the config.

Might be able to bypass this with a bit hacky solution.

@LaaZa LaaZa linked a pull request Mar 27, 2024 that will close this issue
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Add support for InternLM-XComposer2-VL LaaZa/AutoGPTQ
2 participants
@thiner @LaaZa