[go: nahoru, domu]

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.

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

support internlm-xcomposer2d5-7b #1932

Open
wants to merge 6 commits into
base: main
Choose a base branch
from
Open

Conversation

irexyc
Copy link
Collaborator
@irexyc irexyc commented Jul 5, 2024

Motivation

support internlm-xcomposer2d5-7b

related issues #1920

  • base model & image inputs
  • Write & Webpage model (since turbomind doesn't support slora, only one model can be deployed at a time)
  • quantization
  • document
  • video input

@irexyc irexyc added the WIP label Jul 5, 2024
@irexyc irexyc removed the WIP label Jul 7, 2024

## Lora Model

InternLM-XComposer-2.5 trained the LoRA weights for webpage creation and article writing. As TurboMind backend doesn't support slora, only one LoRA model can be deployed at a time, and the LoRA weights need to be merged when deploying the model. LMDeploy provides the corresponding conversion script, which is used as follows:
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Since we support plora, can we support lora directly without merging it?

Copy link
Collaborator Author
@irexyc irexyc Jul 8, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

For specific task, there are actually two LoRA weights(plora + lora) and we can only support one.
https://huggingface.co/internlm/internlm-xcomposer2d5-7b/blob/main/build_mlp.py#L227-L243

@irexyc
Copy link
Collaborator Author
irexyc commented Jul 8, 2024

Currently, the hf model can't be used with transformers api due to InternLM/InternLM-XComposer#354

@lvhan028 lvhan028 added the enhancement New feature or request label Jul 8, 2024
lmdeploy/model.py Outdated Show resolved Hide resolved
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants