-
Notifications
You must be signed in to change notification settings - Fork 29
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
[WIP] support for more vlms #390
base: main
Are you sure you want to change the base?
Conversation
n1ck-guo
commented
Dec 19, 2024
•
edited
Loading
edited
- support for more vlms: solvlm, aria(wip), llama-3.2v-cot, deepseek-vl2
- vl 70b+ on single card
- new processor for all hf-model
- modify get_multimodal_block_names (find language/vision/...)
Signed-off-by: n1ck-guo <[email protected]>
@@ -160,6 +160,9 @@ def __init__( | |||
self.template, model=model, tokenizer=tokenizer, processor=processor, image_processor=image_processor) | |||
dataset = self.template.default_dataset if dataset is None else dataset | |||
|
|||
if model.config.model_type == "deepseek_vl_v2": |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
the setting here is a little tricky. Could the quantizing-non-text-module still be supported?
Signed-off-by: n1ck-guo <[email protected]>
Signed-off-by: n1ck-guo <[email protected]>
any chance to use auto rounder with Aria-UI or Aria? |
|
thanks for the answer. Surely Qwen2VLForConditionalGeneration is not the class to use for Aria-UI-base? Any example or saved model would be great. Thanks |