Re:How to run Qwen 3.5 locally
by Aurornis on Sunday March 08, 2026 @12:58PM
(#47298855)
Long context degradation is a problem with the Qwen3.5 models for me. They have some clever tricks to accelerate attention that favor more recent context.
The models can be frustrating to use if you expect long contexts to behave like they do on SOTA models. In my trials I could give them strict instructions to NOT do something and they would follow it for a short time before ignoring my prompt and doing the things I told it not to do.
The models can be frustrating to use if you expect long contexts to behave like they do on SOTA models. In my trials I could give them strict instructions to NOT do something and they would follow it for a short time before ignoring my prompt and doing the things I told it not to do.