Comment by Foxhuls
11 hours ago
Getting an equivalent amount of ram in GPU's is substantially more expensive than this. A single RTX 6000 Blackwell with 96gb of ram is like $16,000 now. Someone please correct me if I'm wrong but essentially with the Mac Studio you're getting the ability to run larger models, at slower speeds, for a lot cheaper.
Right. Couple things to note though:
- Larger MoE models only need compute proportional to the number of active parameter, so Qwen3.8-Flash-Next (180b with 6b active) runs more tok/sec than Qwen3.8-27b on the same hardware.
- Larger models typically have higher quality output and require fewer thinking tokens to get to an answer.
Consequently, measuring time to complete a task with a wall clock, the speed difference is less than you would expect.
Exactly, optimizing for VRAM where you don’t need the RTX 6000 performance.