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 for deepseek vl2 #401

Merged
merged 5 commits into from
Jan 7, 2025
Merged

support for deepseek vl2 #401

merged 5 commits into from
Jan 7, 2025

Conversation

n1ck-guo
Copy link
Contributor

@n1ck-guo n1ck-guo commented Jan 3, 2025

ut_test:

  • test_support_vlms
  • test_vlms
  • test_2_3bits
  • test_multiple_card_calib
  • test_main_func

Signed-off-by: n1ck-guo <heng.guo@intel.com>
model.forward = model.language.forward
block_names = []
if quant_vision:
block_names.append([f"vision.blocks.{i}" for i in range(len(model.vision.blocks))])
Copy link
Contributor

Choose a reason for hiding this comment

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

better support a arg to pass the vision block name, could be done later

Signed-off-by: n1ck-guo <heng.guo@intel.com>
@wenhuach21
Copy link
Contributor

Please double-check quantizing with the fp_layers; I think it is prone to bugs.

Signed-off-by: n1ck-guo <heng.guo@intel.com>
Signed-off-by: n1ck-guo <heng.guo@intel.com>
Signed-off-by: n1ck-guo <heng.guo@intel.com>
@wenhuach21 wenhuach21 self-requested a review January 7, 2025 01:55
@wenhuach21 wenhuach21 merged commit bf8d68d into main Jan 7, 2025
8 checks passed
@wenhuach21 wenhuach21 deleted the hengguo/fix0103 branch January 7, 2025 01:56
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants