Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> For users with more than 300GB of VRAM, qwen3-coder:480b is also available locally.

I haven't really stayed up on all the AI specific GPUs, but are there really cards with 300GB of VRAM?



You can buy an M3 Ultra Mac Studio and configure it with 512 GB of memory shared between the CPU and the GPU. Will set you back about $9500.


And that'll be two orders of magnitude slower right?


In addition to the already mentioned Apple Mac Studio, NVIDIA sells the GH200 with up to 480GB of VRAM.

My local HPC went for the 120GB version though, but 4 per node.


No, you need multiple GPUs. These models are not intended to be run by the average user.


Not necessarily. You need either multiple GPUs or unified memory. There are a handful of UM platforms out there nowadays (mainly Macs but AMD has some as well albeit none with 300GB ram)


Also the just-released DGX Spark from Nvidia (although it "only" has 128gb of unified memory)


One of its defining features is the ability to link them together at speeds about the same as their ram speed iirc.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: