Skip to main content

Radeon RX 7900 XTX vs NVIDIA GeForce RTX 5080 — Specs & Performance Compared

Verdict: The NVIDIA GeForce RTX 5080 is ahead of the Radeon RX 7900 XTX on all 6 shared benchmarks on file, by 9% to 54%, including 29% in 3DMark Time Spy (32,642 vs 25,214). On qwen2.5:14b, the NVIDIA GeForce RTX 5080 generates 85.0 tokens/s (per Local AI Master) against 62.0 for the Radeon RX 7900 XTX (per Kunal Ganglani), best community run on file for each.

Side-by-side GPU comparison of the Radeon RX 7900 XTX and the NVIDIA GeForce RTX 5080. Released 2022 (Radeon RX 7900 XTX), 2025 (NVIDIA GeForce RTX 5080). MSRP range: $999 · $999. Specs, synthetic benchmarks, AI inference tok/s, and Amazon availability below; each column header links to that part's benchmark page.

vs

*Price sourced from Amazon.com. Price and availability subject to change.

Where to buy

A listing for each part; the current price and stock are on the retailer's page.

Specs side-by-side

SpecRadeon RX 7900 XTXNVIDIA GeForce RTX 5080
vram gb 2416
vram type —GDDR7
tdp w —360
cuda cores —10752

Synthetic benchmarks

Each row says whether a higher or a lower number is better (ranks and times are lower-is-better). Scores come from LanOC Reviews, Wccftech, Digital Trends, 3DMark, HWBOT, HWCooling, Overclocking.com, ComputerBase, TechSpot and Micro Center; each row links its sources, and where they differ the cell shows their median. Empty cells mean we don't yet have that benchmark on file for the SKU.

BenchmarkRadeon RX 7900 XTXNVIDIA GeForce RTX 5080
3DMark Fire Strike (higher is better)
Source: LanOC Reviews		 —79,045 points
3DMark Fire Strike Extreme (higher is better)
Source: LanOC Reviews		 —39,871 points
3DMark Fire Strike Extreme (Graphics) (higher is better)
Source: Wccftech		 37,004 points—
3DMark Fire Strike Ultra (higher is better)
Source: LanOC Reviews		 —21,262 points
3DMark Fire Strike Ultra (Graphics) (higher is better)
Source: Wccftech		 20,587 points—
3DMark Port Royal (higher is better)
Source: Digital Trends, 3DMark, HWBOT, Wccftech, HWCooling		 16,225 points24,924 points
3DMark Speed Way (higher is better)
Source: Overclocking.com, 3DMark, LanOC Reviews, ComputerBase, Wccftech		 6,280 points9,010 points
3DMark Steel Nomad (higher is better)
Source: 3DMark, LanOC Reviews, ComputerBase		 6,862 points8,568 points
3DMark Time Spy (higher is better)
Source: TechSpot, Digital Trends, LanOC Reviews, 3DMark		 25,214 pts32,642 points
3DMark Time Spy (combined) (higher is better)
Source: Micro Center		 27,934 points—
3DMark Time Spy (GPU) (higher is better)
Source: Micro Center		 30,139 points—
3DMark Time Spy (Graphics) (higher is better)
Source: Wccftech		 30,229 points—

AI inference (tokens/sec)

Top community-reported tokens-per-second results for popular LLMs whose weights fit the card's memory, each linked to its source. See the per-SKU benchmark page for the full set including quantization and prompt length.

Radeon RX 7900 XTX

All Radeon RX 7900 XTX AI benchmarks →

NVIDIA GeForce RTX 5080

All NVIDIA GeForce RTX 5080 AI benchmarks →

Quick answers

Which is faster, the Radeon RX 7900 XTX or the NVIDIA GeForce RTX 5080?

The NVIDIA GeForce RTX 5080 is ahead of the Radeon RX 7900 XTX on all 6 shared benchmarks on file, by 9% to 54%, including 29% in 3DMark Time Spy (32,642 vs 25,214).

Which is better for running local LLMs?

On qwen2.5:14b, the NVIDIA GeForce RTX 5080 generates 85.0 tokens/s (per Local AI Master) against 62.0 for the Radeon RX 7900 XTX (per Kunal Ganglani), best community run on file for each.

What did the Radeon RX 7900 XTX and the NVIDIA GeForce RTX 5080 cost at launch?

The Radeon RX 7900 XTX launched at a $999 MSRP and the NVIDIA GeForce RTX 5080 at $999. Street prices drift from launch pricing; the Where to buy section links current listings.

Keep exploring

More guides & deep dives from the SpecPicks archive

Browse all articles & guides →

More reviews from the SpecPicks archive

Browse all reviews →

More buying guides from SpecPicks

Browse all buying guides →

Hardware benchmark data on SpecPicks

All benchmarks →