Repository navigation
Add Qwen and Gemma models to LiteRT - #1192
Open
farook-edev wants to merge 7 commits into
Open
farook-edev wants to merge 7 commits into
farook-edev wants to merge 7 commits into
Conversation
|
MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅ |
Contributor
|
Contributor
Author
|
0.8B Qwen model available here (Converted with GPU patch) |
|
## Problem `ends_with` / `starts_with` (used by `startend:end_checker` and `combination:repeat_prompt`) and `ConstrainedResponse` allow 3 extra characters around the expected text, but measure them with `std::string::size()`, which counts **bytes**. The Python reference uses `len()`, which counts **characters**. Any non-ASCII character in that window makes the C++ check fail where Python passes. Example: `"My answer is no. ⁇ "` is 19 characters (limit 16 + 3, passes in Python) but 21 bytes (fails in C++). ## Proposed Fix Add `utf8_len` (character count) and `utf8_offset` (byte offset of the n-th character) to `common.h`, and use them to size the comparison window in `ends_with` / `starts_with` and the length check in `ConstrainedResponse`. The window is clamped to the string length. ASCII input behaves as before.
Contributor
|
@anhappdev please help put those models to https://mobile.mlcommons-storage.org/app-resources/models/. |
farook-edev
changed the title
Contributor
Author
|
Collaborator
I uploaded to |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR aims to keep Qwen work separate from LiteRT work, it includes all the changes to the pipeline and datasets to allow qwen models (tested only 0.8b) to run on LiteRT backend.