RTX 3060 Ti

Nvidia · 8 GB VRAM

3recorded runs

79.9best tok/s

1sourced

How it compares — 8 GB tier

Best verified tok/s per GPU in this VRAM tier — compare any two like-for-like.

Evidence: 1 sourced run · 2 awaiting review · 1 independent source · first recorded Jul 2, 2026 · last recorded Jul 27, 2026. What these labels mean.

Model	Quant	Peak VRAM	Headroom
InternVL3-8B	Q4_K_M	6.7 GB	1.3 GB
AceReason-Nemotron-1.1-7B	Q4_K_M	6.7 GB	1.3 GB

Peak process memory recorded by the submitter on this card — evidence of what fit, not a guarantee. Plan with the VRAM calculator.

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
RTX 3060 Ti · 8 GB	NVIDIA-Nemotron-Nano-9B-v2	IQ4_XS	llama.cpp llama.cpp b9860	2048	64.2 ✓	1755.9	—	src
RTX 3060 Ti · 8 GB	InternVL3-8B	Q4_K_M	llama.cpp —	512	79.9 …	2963.5	6.7	src
RTX 3060 Ti · 8 GB	AceReason-Nemotron-1.1-7B	Q4_K_M	llama.cpp —	512	79.8 …	2930.7	6.7	src