188GB VRAM on Mac Studio M2 Ultra - EASY
gpu4.pics
via bladeolson26
Apple M2 Ultra
1× Apple M2 Ultra (192GB unified memory) · 192 GB total VRAM · Apple M2 Ultra
192 GB unified Apple
running Dolphin Mixtral 8x7B· Q5
37
tokens / sec
Generation
37tok/s
higher is better
Prompt / prefill
—tok/s
higher is better
Time to first token
—
lower is better
Composite score
—
Model size
7B
Runtime
—
All benchmarks on this rig
| Model | Quant | Gen | Prompt | TTFT | Power | Perf/W | $/Mtok | Score |
|---|---|---|---|---|---|---|---|---|
| Dolphin Mixtral 8x7B | Q5 | 37 | — | — | — | — | — | — |
| Goliath 120B | Q4 | 7 | — | — | — | — | — | — |