GPU comparison
AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) vs NVIDIA RTX 4090 (24GB) 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.
| AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) | NVIDIA RTX 4090 (24GB) | |
|---|---|---|
| Memory | 128.0 GB | 24.0 GB |
| Bandwidth | 256 GB/s | 1008 GB/s |
| Price (approx) | $2,999 | $2,999 |
| LLMs it runs | 57 of 64 | 47 of 64 |
| Best model it runs | Qwen3.8 Flash Next (MoE) · 45–67 tok/s | Qwen3.8 27B · 38–57 tok/s |
The AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) runs 10 more of the tracked models (57 vs 47), thanks to its 128.0 GB of memory. The NVIDIA RTX 4090 (24GB) has more memory bandwidth (1008 GB/s), so it generates tokens faster at the same model and quant.