Has any individual/SMB purchased this yet? Aware the ROI isn't there for most individuals that don't heavily use LLM for coding or multi agentic workflows -- but for those that it does make sense this is ~$68k cheaper than the next alternative making this the most realistic option for consumer grade access to frontier models.
Specifically interested if anyone has ran GLM-5.3 (753B MoE) in 4-bit on a 512GB Mac Studio. Looking for real world usage numbers and/or any benchmarks you might have performed.
Thanks!
bigyabai•35m ago
GLM-5.3 would be bottlenecked by the M5's GPU, here. I don't have any hard numbers, but Apple Silicon is known for weak prefill performance on larger models (yes, even MoEs).
immutlex•41m ago
Specifically interested if anyone has ran GLM-5.3 (753B MoE) in 4-bit on a 512GB Mac Studio. Looking for real world usage numbers and/or any benchmarks you might have performed.
Thanks!
bigyabai•35m ago