GPU comparison
AMD Ryzen AI Max+ 395 “Strix Halo” (64GB) vs NVIDIA RTX 4080 (16GB) 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 4080 (16GB) | |
|---|---|---|
| Memory | 64.0 GB | 16.0 GB |
| Bandwidth | 256 GB/s | 717 GB/s |
| Price (approx) | $1,999 | $1,549 |
| LLMs it runs | 26 of 34 | 18 of 34 |
| Best model it runs | Gemma 4 31B · 10–15 tok/s | Gemma 4 26B-A4B (MoE) · 141–212 tok/s |
The AMD Ryzen AI Max+ 395 “Strix Halo” (64GB) runs 8 more of the tracked models (26 vs 18), thanks to its 64.0 GB of memory. The NVIDIA RTX 4080 (16GB) has more memory bandwidth (717 GB/s), so it generates tokens faster at the same model and quant.