GPU comparison
Mac Studio M5 Ultra (256GB) vs NVIDIA RTX 5090 (32GB) 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 M5 Ultra (256GB) | NVIDIA RTX 5090 (32GB) | |
|---|---|---|
| Memory | 256.0 GB | 32.0 GB |
| Bandwidth | 1228 GB/s | 1792 GB/s |
| Price (approx) | $9,499 | $4,399 |
| LLMs it runs | 60 of 64 | 48 of 64 |
| Best model it runs | GLM-5.3 Flash (MoE) · 61–91 tok/s | Qwen3.8 27B · 48–72 tok/s |
The Mac Studio M5 Ultra (256GB) runs 12 more of the tracked models (60 vs 48), thanks to its 256.0 GB of memory. The NVIDIA RTX 5090 (32GB) has more memory bandwidth (1792 GB/s), so it generates tokens faster at the same model and quant.