Base Model?

#2
by Downtown-Case - opened

Is a base model for Qwen3 32B not available yet?

Is a base model for Qwen3 32B not available yet?

good catch we're gonna upload and update soon

Edit: nevermind Qwen never released base models for those variants

Doesn't look like the Qwen team are releasing the two largest base models

Is there any confirmation of that? Even a low key one?

MS-SWIFT repo lists support for qwen3 32b base and 235b-a22b base.

MS Swift is developed by ModelScope. ModelScope is developed by Alibaba Cloud, which is the same company as the one releasing Qwen series.

So, unless they have internal miscommunication, it's likely that we'll see base models for 32b and 235B!

https://github.com/modelscope/ms-swift/commit/ad907793e9d3450bf1b28ac0e1eebb5f4890265b

@adamo1139 Nice find!

The modelscope link is dead though, unfortunately.

Any word on this?

Oh apologies, we actually didnt know Qwen didnt release base models for them, so unfortunately until they officially release it, we cannot do anything about it

Your need to confirm your account before you can post a new comment.

Sign up or log in to comment