GPU comparison
NVIDIA RTX 4070 (12GB) vs NVIDIA RTX 4080 (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 4070 (12GB) | NVIDIA RTX 4080 (16GB) | |
|---|---|---|
| Memory | 12.0 GB | 16.0 GB |
| Bandwidth | 504 GB/s | 717 GB/s |
| Price (approx) | $549 | $1,549 |
| LLMs it runs | 19 of 48 | 27 of 48 |
| Best model it runs | Devstral Small 2 24B · 35–53 tok/s | Gemma 4 26B-A4B (MoE) · 141–212 tok/s |
The NVIDIA RTX 4080 (16GB) runs 8 more of the tracked models (27 vs 19), thanks to its 16.0 GB of memory. The NVIDIA RTX 4080 (16GB) has more memory bandwidth (717 GB/s), so it generates tokens faster at the same model and quant.