-
Notifications
You must be signed in to change notification settings - Fork 54
Apply Qwen3 PR from llama.cpp #355
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
Thanks! I was just in the process of doing the same. Does |
@@ -331,6 +331,8 @@ class MODEL_TENSOR(IntEnum): | |||
MODEL_ARCH.QWEN: "qwen", | |||
MODEL_ARCH.QWEN2: "qwen2", | |||
MODEL_ARCH.QWEN2MOE: "qwen2moe", | |||
MODEL_ARCH.QWEN3: "qwen3", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You are missing the QWEN3
and QWEN3MOE
enum entries further up in class MODEL_ARCH(IntEnum)
OK, I'll merge this and will add the missing enum entries separately. |
Ok - my other concern was that |
I've only tested that the MOE works.
|
And a no-think example
|
I also tested before merging and it seemed to be working correctly. |
I've just ported over the Qwen3 PR. So it is missing the layers/model type, and does not have tests, etc.