Spaces:
Running
on
CPU Upgrade
How to submit Qwen model?
Hi, we hope to submit our model trained from Qwen-72b, but since it has not been integrated into transformers, it requires the use of use_remote_code=True. Recently, we noticed that Qwen-72B appeared in the leaderboard. So, how can i manage to submit my model?
Besides, why does the Qwen-72B page disappeared?
Hi!
At the moment, it's not possible to submit Qwen fine-tunes. We manually ran Qwen-72B as we feel it's important for the community to know the performance of pretrained models, that they will later on be able to use. However, as soon as Qwen is in a stable version of transformers, we will allow submissions on the leaderboard.
Regarding the details page, I suspect that they were not pushed to the hub as they should have. cc @SaylorTwift
Hi, we had an issue uploading the details dataset, I just re-uploaded it.
Would you please also, manually run Chat ver. of Qwen-72B as well?