GPU comparison
NVIDIA RTX 6000 Ada (48GB) vs AMD RX 9070 (16GB) for local LLMs
How the two stack up for running open-source LLMs locally — memory, bandwidth, price, and how many of the 48 tracked models each one can run.
| NVIDIA RTX 6000 Ada (48GB) | AMD RX 9070 (16GB) | |
|---|---|---|
| Memory | 48.0 GB | 16.0 GB |
| Bandwidth | 960 GB/s | 645 GB/s |
| Price (approx) | $6,800 | $569 |
| LLMs it runs | 37 of 48 | 27 of 48 |
| Best model it runs | Gemma 4 31B · 20–30 tok/s | Gemma 4 26B-A4B (MoE) · 132–198 tok/s |
The NVIDIA RTX 6000 Ada (48GB) runs 10 more of the tracked models (37 vs 27), thanks to its 48.0 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.