Home › GPUs & Macs › NVIDIA RTX 3080 Ti vs NVIDIA RTX 3090
NVIDIA RTX 3080 Ti vs NVIDIA RTX 3090 for local LLMs
Across 3 tests measured on both, the NVIDIA RTX 3090 is faster in 2 and generates a median 1.06× the tokens/s of the NVIDIA RTX 3080 Ti.
On Llama 3.1 8B Instruct (Q4_K_M, LocalScore) the NVIDIA RTX 3080 Ti generates 90.5 tokens/s and the NVIDIA RTX 3090 96.3 tokens/s (measured).
Same model, same quant, same test
| Model | Quant / test | RTX 3080 Ti tok/s | RTX 3090 tok/s | Faster | Sources and run details |
|---|---|---|---|---|---|
| Llama 3.1 8B Instruct | Q4_K_M LocalScore | 90.5 llamafile | 96.3 llamafile | RTX 3090 +6% | LocalScore RTX 3080 Ti · LocalScore RTX 3090 |
| Llama 3.2 1B Instruct | Q4_K_M LocalScore | 313 llamafile | 270 llamafile | RTX 3080 Ti +16% | LocalScore RTX 3080 Ti · LocalScore RTX 3090 |
| Qwen2.5 14B Instruct | Q4_K_M LocalScore | 46.1 llamafile | 56.2 llamafile | RTX 3090 +22% | LocalScore RTX 3080 Ti · LocalScore RTX 3090 |
Generation speed in tokens/s, best measured backend for each chip (shown under the number). Backends can differ between the two chips; check them before drawing conclusions. Measured results only, nothing estimated.
Specs
| NVIDIA RTX 3080 Ti | NVIDIA RTX 3090 | |
|---|---|---|
| Memory | 12 GB | 24 GB |
| Memory bandwidth | 912 GB/s | 936 GB/s |
All results for the NVIDIA RTX 3080 Ti · All results for the NVIDIA RTX 3090 · More comparisons