Home › GPUs & Macs › NVIDIA RTX 3060 vs NVIDIA Tesla P40
NVIDIA RTX 3060 vs NVIDIA Tesla P40 for local LLMs
Across 3 tests measured on both, the NVIDIA RTX 3060 is faster in 3 and generates a median 1.73× the tokens/s of the NVIDIA Tesla P40.
On Llama 2 7B (Q4_0, llama-bench) the NVIDIA RTX 3060 generates 80.6 tokens/s and the NVIDIA Tesla P40 59.4 tokens/s (measured).
Same model, same quant, same test
| Model | Quant / test | RTX 3060 tok/s | Tesla P40 tok/s | Faster | Sources and run details |
|---|---|---|---|---|---|
| Llama 2 7B | Q4_0 llama-bench | 80.6 Vulkan | 59.4 Vulkan | RTX 3060 +36% | llama.cpp #10879 RTX 3060 · llama.cpp #10879 Tesla P40 |
| Llama 3.2 1B Instruct | Q4_K_M LocalScore | 185 llamafile | 105 llamafile | RTX 3060 +77% | LocalScore RTX 3060 · LocalScore Tesla P40 |
| Qwen2.5 14B Instruct | Q4_K_M LocalScore | 26.6 llamafile | 15.4 llamafile | RTX 3060 +73% | LocalScore RTX 3060 · LocalScore Tesla P40 |
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 3060 | NVIDIA Tesla P40 | |
|---|---|---|
| Memory | 12 GB | 24 GB |
| Memory bandwidth | 360 GB/s | 346 GB/s |
All results for the NVIDIA RTX 3060 · All results for the NVIDIA Tesla P40 · More comparisons