Intel b70 is going for $1300 and should I just pull the trigger?
retained text
I have an M1 Max with 64GB RAM. It can fit some large models, which is great, but the performance isn’t really where I want it to be. A lot of the bigger models are technically usable, but pretty slow. I was originally planning to upgrade to a 128GB MacBook, but the problem is my current Mac already does everything I need for software development. There’s really nothing wrong with it besides local LLM performance, so spending a bunch of money just to replace it feels kind of wasteful. I could afford an M4/M5 Max with 128GB, or instead put that money toward something like four of these b70 GPUs. I’m still pretty new to the local LLM hardware side of things though, and I’m having a hard time figuring out what actually makes sense. I’ve been asking ChatGPT a lot, but it keeps giving me conflicting or straight-up hallucinated answers, so I’d rather hear from people who actually run these setups. I’m pretty open to recommendations. I mainly want to get deeper into AI and local models and not fall behind. I already work in software, so it feels like this is something worth investing some time and money into.
document fetched 2026-09-21 00:36 · canonical https://reddit.com/r/LocalLLaMA/comments/1wlnsv3/intel_b70_is_going_for_1300_and_should_i_just