NVIDIA GeForce RTX 3070 Ti — Benchmarks & Specs

Bottom line: how fast is the NVIDIA GeForce RTX 3070 Ti?

SpecPicks median: 80.6 tokens/sec across 6 cited single-user runs of 7-9B dense models at 4-bit (Q4) under llama.cpp (range 33.9–102 tok/s), from GPU Battle, LocalScore.ai (2 runs), TechReviewer, Kubito (2 runs). At the $740.00 Amazon price of the cheapest linked listing (checked 2026-10-21 04:13 UTC; prices change), that is a SpecPicks Value Index of 108.9 tok/s per $1,000. At 1440p Ultra, the NVIDIA GeForce RTX 3070 Ti averages 67 fps in Tom's Hardware GPU Hierarchy — rasterization suite geomean, per Tom's Hardware. For local LLM inference it generates 59.4 tokens/sec running qwen3.5:9b at q4_K_M under ollama, per Patrick Hughes. In 3DMark Fire Strike it scores 26,516 points, per Digital Trends. Its 8 GB of VRAM is the binding constraint for local inference: the qwen3.5:9b run above (9B parameters) is the largest model on file at 4-bit quantization on this card.

Every figure above is a row in the tables below, and each row links out to the review or public benchmark database the number was taken from; the SpecPicks median is computed here from the runs it names, each a row in the AI table, and the per-$1,000 figure divides it by the dated listing price. SpecPicks aggregates published measurements; it does not report first-party benchmark runs.