GPUs & Macs › NVIDIA GTX 1660 Ti

NVIDIA GTX 1660 Ti: local LLM speed

1 measured result for the NVIDIA GTX 1660 Ti.

Measured results

Model	Generate tok/s	Prompt tok/s	Quant	Backend	Test	Source
Llama 3.2 1B Instruct	37.1	829	Q4_K_M	llamafile	LocalScore avg	LocalScore details

Vendor	NVIDIA
Memory	6 GB
Memory bandwidth	288 GB/s
Spec source	Wikipedia (List of Nvidia graphics processing units)

As an Amazon Associate I earn from qualifying purchases.

Affiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.

No measured run yet; estimates use spec bandwidth × 0.66 (typical efficiency for NVIDIA). Fit assumes the whole model in GPU memory with ~8K context.

Model	Best quant that fits	File	Expected tok/s
gemma-4-12B-it	UD-IQ2_M	4.2 GB	44.8 estimate
Qwen3.5-9B	UD-IQ3_XXS	4.0 GB	47.0 estimate
Qwen3.5-4B	Q8_0	4.5 GB	42.1 estimate
MiMo-V2.6-Distill-Qwen-9B	Q3_K_M	4.5 GB	42.2 estimate
gemma-4-E4B-it	Q3_K_M	4.1 GB	46.5 estimate
Qwen3-4B	Q8_0	4.3 GB	44.1 estimate
gemma-3-1b-it	Q8_0	1.1 GB	157 estimate
gemma-4-E2B-it	Q4_0	2.8 GB	66.5 estimate
DeepSeek-R1-0528-Qwen3-8B	Q3_K_M	4.1 GB	45.8 estimate
Llama-3.2-3B-Instruct	Q8_0	3.4 GB	55.2 estimate
Phi-4-mini-instruct	Q8_0	4.1 GB	46.2 estimate
gemma-3-4b-it	Q8_0	4.1 GB	45.7 estimate
Qwen3-14B	UD-IQ1_M	4.1 GB	46.5 estimate
gemma-3-12b-it	UD-IQ2_M	4.4 GB	43.3 estimate