GPU comparison
AMD Ryzen AI Max+ 395 “Strix Halo” (64GB) 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 34 tracked models each one can run.
| AMD Ryzen AI Max+ 395 “Strix Halo” (64GB) | NVIDIA RTX 6000 Ada (48GB) | |
|---|---|---|
| Memory | 64.0 GB | 48.0 GB |
| Bandwidth | 256 GB/s | 960 GB/s |
| Price (approx) | $1,999 | $6,800 |
| LLMs it runs | 26 of 34 | 26 of 34 |
| Best model it runs | Gemma 4 31B · 10–15 tok/s | Gemma 4 31B · 20–30 tok/s |
Both run the same number of models, but the AMD Ryzen AI Max+ 395 “Strix Halo” (64GB) has more memory (64.0 GB), so it can hold larger models at higher quality. The NVIDIA RTX 6000 Ada (48GB) has more memory bandwidth (960 GB/s), so it generates tokens faster at the same model and quant.