Qwen 3.8 Flash Next locally on simple mobile phone at 3.5 tok/s
Qwen 3.8 Flash Next (80gb) now at 3.5 tok/s on 12gb mid range android phone thanks to some optimizations and with a low quantization on dense part. I don't want to promote the project, but simply show that it's possible on a $400–$500 phone
评论
?
参与讨论