News for hackers. Stuff that matters.

A comment on Gemma 4 12B: A unified, encoder-free multimodal model. Parent comment

Re:Gemma 4 12B: A unified, encoder-free multimodal model by sgt101 on Thursday June 04, 2026 @03:46PM (#48403691)
31B won't run in 48GB for me - it needs 54.
[ Parent | on HN ]
Re:Gemma 4 12B: A unified, encoder-free multimodal model by yassa9 on Thursday June 04, 2026 @04:51PM (#48404456)
what quantization did u try ? u can use Q4 quantization, im pretty sure that 48GB would be enough
[ Parent | on HN ]
Re:Gemma 4 12B: A unified, encoder-free multimodal model by sgt101 on Saturday June 06, 2026 @03:05AM (#48422220)
8bits is fine.... I was talking full bore.
[ Parent | on HN ]