Home › GPUs & Macs › NVIDIA RTX 4060 Ti vs NVIDIA RTX A2000
NVIDIA RTX 4060 Ti vs NVIDIA RTX A2000 for local LLMs
Across 3 tests measured on both, the NVIDIA RTX 4060 Ti is faster in 3 and generates a median 1.43× the tokens/s of the NVIDIA RTX A2000.
On Llama 2 7B (Q4_0, llama-bench) the NVIDIA RTX 4060 Ti generates 64.0 tokens/s and the NVIDIA RTX A2000 45.7 tokens/s (measured).
Same model, same quant, same test
| Model | Quant / test | RTX 4060 Ti tok/s | RTX A2000 tok/s | Faster | Sources and run details |
|---|---|---|---|---|---|
| Llama 2 7B | Q4_0 llama-bench | 64.0 CUDA | 45.7 Vulkan | RTX 4060 Ti +40% | llama.cpp #15013 RTX 4060 Ti · llama.cpp #10879 RTX A2000 |
| Llama 3.1 8B Instruct | Q4_K_M LocalScore | 43.3 llamafile | 28.7 llamafile | RTX 4060 Ti +51% | LocalScore RTX 4060 Ti · LocalScore RTX A2000 |
| Llama 3.2 1B Instruct | Q4_K_M LocalScore | 167 llamafile | 117 llamafile | RTX 4060 Ti +43% | LocalScore RTX 4060 Ti · LocalScore RTX A2000 |
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 4060 Ti | NVIDIA RTX A2000 | |
|---|---|---|
| Memory | 8 GB | 6 GB |
| Memory bandwidth | 288 GB/s | 288 GB/s |
All results for the NVIDIA RTX 4060 Ti · All results for the NVIDIA RTX A2000 · More comparisons