GPU comparison
NVIDIA RTX 5070 (12GB) vs NVIDIA RTX 5090 (32GB) 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.
| NVIDIA RTX 5070 (12GB) | NVIDIA RTX 5090 (32GB) | |
|---|---|---|
| Memory | 12.0 GB | 32.0 GB |
| Bandwidth | 672 GB/s | 1792 GB/s |
| Price (approx) | $629 | $4,399 |
| LLMs it runs | 13 of 34 | 25 of 34 |
| Best model it runs | Devstral Small 2 24B · 45–68 tok/s | Gemma 4 31B · 50–75 tok/s |
The NVIDIA RTX 5090 (32GB) runs 12 more of the tracked models (25 vs 13), thanks to its 32.0 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.