Phi3 runner uses TextLLMRunner#11551
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/11551
Note: Links to docs will display an error until the docs builds have been completed. ❌ 27 New FailuresAs of commit 7418d2e with merge base 042eb1a ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This PR needs a
|
f240099 to
277fd57
Compare
|
@larryliu0820 do you plan to have this as part of 0.7? |
|
ok rebasing |
277fd57 to
50a8b60
Compare
As titled, this PR started to use `TextLLMRunner` to run phi-3-mini. Eager model comes from Huggingface, not using kv cache as custom op because it is only being supported on Optimum-executorch repo. Performance may not be the best.
2f55b06 to
7418d2e
Compare
|
Merged in #12482 |
As titled.