ipex-llm icon indicating copy to clipboard operation
ipex-llm copied to clipboard
verified publisher icon intel-analytics

Feature Request: Updated finetuning support

Open electroglyph opened this issue 9 months ago • 1 comments

This repo is called "Accelerate local LLM inference and finetuning..." but the finetuning has been neglected for a long time. Axolotl v0.4.0 is 15+ months old now. Humbly requesting some work in this area. Thanks!

electroglyph avatar Apr 10 '25 05:04 electroglyph

Hi @0xDEADFED5

The finetuning feature was built based on customers' requirements. So, keeping it aligned with the latest version is not the first priority.

Anyway, if you are trying to use this feature and find gaps (model or libs gap). Please let us know. :)

qiyuangong avatar Apr 14 '25 08:04 qiyuangong