GPU comparison
Mac Studio M3 Ultra (96GB) vs NVIDIA RTX 6000 Ada (48GB) for local LLMs
How the two stack up for running open-source LLMs locally — memory, bandwidth, price, and how many of the 64 tracked models each one can run.
| Mac Studio M3 Ultra (96GB) | NVIDIA RTX 6000 Ada (48GB) | |
|---|---|---|
| Memory | 96.0 GB | 48.0 GB |
| Bandwidth | 819 GB/s | 960 GB/s |
| Price (approx) | $5,299 | $6,800 |
| LLMs it runs | 55 of 64 | 52 of 64 |
| Best model it runs | Qwen3.8 27B · 17–25 tok/s | Qwen3.8 27B · 23–34 tok/s |
The Mac Studio M3 Ultra (96GB) runs 3 more of the tracked models (55 vs 52), thanks to its 96.0 GB of memory. The NVIDIA RTX 6000 Ada (48GB) has more memory bandwidth (960 GB/s), so it generates tokens faster at the same model and quant.