GPU comparison
NVIDIA RTX 4090 (24GB) vs AMD RX 9070 XT (16GB) 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.
| NVIDIA RTX 4090 (24GB) | AMD RX 9070 XT (16GB) | |
|---|---|---|
| Memory | 24.0 GB | 16.0 GB |
| Bandwidth | 1008 GB/s | 645 GB/s |
| Price (approx) | $2,999 | $699 |
| LLMs it runs | 47 of 64 | 45 of 64 |
| Best model it runs | Qwen3.8 27B · 38–57 tok/s | Qwen3.8 27B · 38–58 tok/s |
The NVIDIA RTX 4090 (24GB) runs 2 more of the tracked models (47 vs 45), thanks to its 24.0 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.