GPUs & Macs › Nvidia GTX 1660 Ti Laptop
Nvidia GTX 1660 Ti Laptop: local LLM speed
Llama 2 7B (Q4_0) generates 57.3 tokens/s on the Nvidia GTX 1660 Ti Laptop (Vulkan), and processes prompts at 514 tokens/s.
Measured results
| Model | Generate tok/s | Prompt tok/s | Quant | Backend | Test | Source |
|---|---|---|---|---|---|---|
| Llama 2 7B | 57.3 | 514 | Q4_0 | Vulkan FA on | tg128 | llama.cpp #10879 details |
| Llama 2 7B | 56.6 | 512 | Q4_0 | Vulkan FA off | tg128 | llama.cpp #10879 details |
What fits on the Nvidia GTX 1660 Ti Laptop
Estimates for this chip are scaled from a measured run: Llama 2 7B Q4_0 at 57.3 tok/s (Vulkan, source). Fit assumes the whole model in GPU memory with ~8K context.
Memory size unknown for this device.
Similar hardware
NVIDIA RTX 3050 Laptop · NVIDIA RTX 3050 Ti Laptop · NVIDIA RTX A1000 Laptop · NVIDIA RTX A2000 Laptop · NVIDIA RTX 500 Ada Laptop · NVIDIA T1200 Laptop · NVIDIA RTX 4050 Laptop · NVIDIA RTX 3060 Laptop