Qwen3.8 27B scores 52 on Artificial Analysis

Re:Qwen3.8 27B scores 52 on Artificial Analysis by bertili on Monday August 17, 2026 @02:24PM (#49335455)

I can't shake this the existential feeling that this compact series of 27G bytes represent something profound and universal.

[ Parent | on HN ]

Re:Qwen3.8 27B scores 52 on Artificial Analysis by kzrdude on Monday August 17, 2026 @03:02PM (#49335985)

Raw model size is around 27x2 GB since it's in BF16 format

[ Parent | on HN ]

Re:Qwen3.8 27B scores 52 on Artificial Analysis by macwhisperer on Monday August 17, 2026 @05:21PM (#49337822)

4-bit quant sits at 15.72GB

[ Parent | on HN ]