etLLM changing model export based on how it's named is very confusing
Open
@jackzhxng is already working on this.
Since Dec 11, 2025.
- Dominant language
- Python
- Stars
- 5k
- Forks
- 1.2k
- Avg merge
- 2d 10h
- Merged PRs (30d)
- 581
Description
🚀 The feature, motivation and pitch
etLLM has logic that says if "int8" is in the checkpoint name, we should quantize it with int8 weight-only quantization. This is extremely confusing https://github.com/pytorch/executorch/blob/fae5d1b53df0a2a9551711c9f4529b5bfcc7ba77/examples/models/llama/model.py#L167-L170.
Someone recently tried to export qwen that had path name "/home/datta0/models/Qwen3-0.6B-int8-int4-unsloth-torchao/pytorch_model_converted.bin" and ExecuTorch tries to apply weight-only quantization.
Alternatives
No response
Additional context
No response
RFC (Optional)
No response
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.