Skip to content

Conversation

abhishek-singh591
Copy link

@abhishek-singh591 abhishek-singh591 commented Sep 10, 2025

This PR resolves an issue encountered when executing the model via CLI, where an error was raised due to the absence of the custom_IO.yaml file. In previous versions, the CLI did not explicitly generate a separate configuration for custom I/O. To address this, a script has been added that automatically creates the custom_IO.yaml file based on the specified modeling class.

Additionally, an optional argument --mxint8_kv_cache has been introduced in QEfficient.cloud.export(). This flag enables compression of the Present/Past KV cache to MXINT8 format using the generated CustomIO configuration.

Signed-off-by: abhishek-singh591 <sabhis@qti.qualcomm.com>
Signed-off-by: abhishek-singh591 <sabhis@qti.qualcomm.com>
Signed-off-by: abhishek-singh591 <sabhis@qti.qualcomm.com>
Signed-off-by: abhishek-singh591 <sabhis@qti.qualcomm.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants