Add ONNX external data support for models >2GB #3844
Closed
+968
−730
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
@SinanGncgl , thanks for taking this up. However, I am working on a major refactoring of onnx-ir that would affect your PR. The refactor also covers needed infrastructure #3596 This PR would need to wait till the refactor is complete. Sorry if this wasn't communicated earlier. |
No worries I can mark it as draft, thanks for letting me know. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Pull Request Template
Checklist
cargo run-checkscommand has been executed.Related Issues/PRs
#3812
Changes
Problem: ONNX models >2GB cannot be loaded due to protobuf size limits. The protobuf format stores tensor data inline, causing failures for large models.
Solution: Added external data support where tensor weights are stored in separate .bin files. The .onnx file contains only metadata (location, offset, length, checksum). This enables loading models of any size while maintaining full backward compatibility.
Testing
Tested with small units which has 2.68GB of model.