GPU comparison
NVIDIA RTX 5090 (32GB) 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 48 tracked models each one can run.
| NVIDIA RTX 5090 (32GB) | NVIDIA RTX 6000 Ada (48GB) | |
|---|---|---|
| Memory | 32.0 GB | 48.0 GB |
| Bandwidth | 1792 GB/s | 960 GB/s |
| Price (approx) | $4,399 | $6,800 |
| LLMs it runs | 35 of 48 | 37 of 48 |
| Best model it runs | Gemma 4 31B · 50–75 tok/s | Gemma 4 31B · 20–30 tok/s |
The NVIDIA RTX 6000 Ada (48GB) runs 2 more of the tracked models (37 vs 35), thanks to its 48.0 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.