Qualcomm AI Engine Direct - PTQ for llama3.2 1b/3b #12700
Merged
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/12700
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New Failures, 8 Pending, 1 Unrelated FailureAs of commit fd7ee71 with merge base a7eefd0 ( NEW FAILURES - The following jobs have failed:
BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
haowhsu-quic
force-pushed
the
dev_seqmse
branch
from
7481013 to
3f2cad4
Compare
haowhsu-quic
changed the title
|
@pytorchbot label "release notes: qualcomm" |
pytorch-bot
bot
added
the
release notes: qualcomm
|
Hi @cccclai, this PR is for introducing SeqMSE for PTQ llama3.2_1/3b_instruct / qwen3 0.6b. Please have a look, thank you. |
|
@haowhsu-quic can you rebase? I'm having issues to import. Also feel free to tag me more often this week, trying to have PRs landed asap |
- add ptq recipe for llama3.2 1b/3b - add seq_mse support for helping quantizing 1b model - complement qnn_llama_runner for smollm2
haowhsu-quic
force-pushed
the
dev_seqmse
branch
from
429d4d9 to
e5cd418
Compare
|
Hi @cccclai, please help merge this when you're available. Thank you. |
cccclai
approved these changes
Sep 11, 2025
swolchok
reviewed
Sep 11, 2025
Comment on lines
+185
to
+186
| def __init__(self): | ||
| super(RemoveSeqMse, self).__init__() |
StrycekSimon
pushed a commit
to nxp-upstream/executorch
that referenced
this pull request
Sep 23, 2025
### Summary - add ptq recipe for llama3.2 1b/3b - add seq_mse support for helping quantizing 1b model - complement qnn_llama_runner for smollm2 ### Test Plan ```bash python examples/qualcomm/oss_scripts/llama/llama.py -b build-android -H $HOST -s $SN -m SM8750 --temperature 0 --model_mode hybrid --prefill_ar_len 128 --max_seq_len 1024 --decoder_model llama3_2-1b_instruct --prompt "I would like to learn python, could you teach me with a simple example?" --artifact ./llama_artifact --tasks wikitext --limit 1 --compile_only --params ../.llama/checkpoints/Llama3.2-1B-Instruct/params.json --tokenizer_model ../.llama/checkpoints/Llama3.2-1B-Instruct/tokenizer.model --checkpoint ../.llama/checkpoints/Llama3.2-1B-Instruct/consolidated.00.pth ```
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.
Summary
Test Plan