-
Notifications
You must be signed in to change notification settings - Fork 234
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
Download fix #1366
Download fix #1366
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/torchchat/1366
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ✅ No FailuresAs of commit 5747a71 with merge base 5da240a ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
a74ddc9
to
0c14af5
Compare
Downloading a Mistral model fails because it includes multiple weight mapping files. The regression was introduced in commit `766bee9f4a1fcb187fae543a525495d3ff482097`. I'm unclear on the original intent, but perhaps the exception was meant to apply only to Granite models. This isn’t an ideal fix, but it does enable Mistral to be downloaded and used for chat. Signed-off-by: Sébastien Han <seb@redhat.com>
The previous logic didn't handle .bin files, so if a model (like mistral) has both .bin and .safetensors, it would download both. Branch: download-fix Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
This will not actually be needed for mistral with the fix in download to handle .bin files, but it may be needed for other models, so it's worth having. Branch: download-fix Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
9a0b5fa
to
5747a71
Compare
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.
Looks solid to me, thanks for the follow up
@Gasoonjia @vmpuri Can one of yall sanity test then ship it?
* fix: allow multiple weight mapping files for mistral Downloading a Mistral model fails because it includes multiple weight mapping files. The regression was introduced in commit `766bee9f4a1fcb187fae543a525495d3ff482097`. I'm unclear on the original intent, but perhaps the exception was meant to apply only to Granite models. This isn’t an ideal fix, but it does enable Mistral to be downloaded and used for chat. Signed-off-by: Sébastien Han <seb@redhat.com> * fix(download): Fix safetensors/bin/pth download logic The previous logic didn't handle .bin files, so if a model (like mistral) has both .bin and .safetensors, it would download both. Branch: download-fix Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> * fix(convert hf): Better logic to handle multiple weight mapping files This will not actually be needed for mistral with the fix in download to handle .bin files, but it may be needed for other models, so it's worth having. Branch: download-fix Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> --------- Signed-off-by: Sébastien Han <seb@redhat.com> Signed-off-by: Gabe Goodhart <ghart@us.ibm.com> Co-authored-by: Sébastien Han <seb@redhat.com>
Closes #1360
Replaces #1346
This is an updated fix for the root cause of the failure to download Mistral that was introduced in #1255