NVIDIA GeForce RTX 5090 vs NVIDIA GeForce RTX 4090 — Specs & Performance Compared

Side-by-side comparison of 2 GPU parts: NVIDIA GeForce RTX 5090 vs NVIDIA GeForce RTX 4090. Release years: 2022, 2025. MSRP range: $1,999 · $1,599. Specs, synthetic benchmarks, AI inference tok/s, and Amazon availability — click into any column header for the per-SKU benchmark page.

vs

*Price sourced from Amazon.com. Last updated 2026-05-27. Price and availability subject to change.

Specs side-by-side

SpecNVIDIA GeForce RTX 5090NVIDIA GeForce RTX 4090
vram gb 3224
vram type GDDR7GDDR6X
tdp w 575450
cuda cores 2176016384

Synthetic benchmarks

Higher is better. Per-row scores come from public benchmark databases (PassMark, TechPowerUp, Geekbench, Cinebench). Empty cells mean we don't yet have that benchmark on file for the SKU.

BenchmarkNVIDIA GeForce RTX 5090NVIDIA GeForce RTX 4090
3DMark Port Royal 36,667 points—
3DMark Speed Way 14,444 points—
3DMark Steel Nomad (4K) 14,133 points—
3DMark Time Spy 32,500 pts—
3DMark Time Spy Extreme —38,450 pts
3DMark Time Spy Extreme (Graphics) —19,000 pts
3DMark Time Spy GPU —35,200 pts
Deep Learning Benchmark (ResNet-50) —128,000 images/sec
PassMark G2D Mark 1,412 pts1,302 pts
PassMark G3D Mark 38,935 pts38,066 pts
Tom's Hardware GPU Hierarchy 2 rank2 rank

AI inference (tokens/sec)

Top community-reported tokens-per-second results for popular LLMs. See the per-SKU benchmark page for the full set including quantization, prompt length, and source citations.

NVIDIA GeForce RTX 5090

  • Qwen2.5-Coder-7B-Instruct: 5841.0 tok/s
  • Llama 3.1 8B FP8: 420.0 tok/s
  • Llama 2 7B: 263.6 tok/s

All NVIDIA GeForce RTX 5090 AI benchmarks →

NVIDIA GeForce RTX 4090

  • llama3:8b: 440.0 tok/s
  • Llama 2 7B: 194.0 tok/s
  • Llama 3.1 8B: 165.0 tok/s

All NVIDIA GeForce RTX 4090 AI benchmarks →

Keep exploring