Radeon RX 9070 XT vs NVIDIA GeForce RTX 5090 — Specs & Performance Compared
Verdict: The NVIDIA GeForce RTX 5090 is ahead of the Radeon RX 9070 XT on all 5 shared benchmarks on file, by 7% to 147%, including 147% in 3DMark Speed Way (15,488 vs 6,267). On gpt-oss:20b, the NVIDIA GeForce RTX 5090 generates 298.2 tokens/s (per Hardware Corner) against 91.9 for the Radeon RX 9070 XT (per hirokuze (GitHub)), best community run on file for each.
Side-by-side GPU comparison of the Radeon RX 9070 XT and the NVIDIA GeForce RTX 5090. Both released in 2025. MSRP range: $599 · $1,999. Specs, synthetic benchmarks, AI inference tok/s, and Amazon availability below; each column header links to that part's benchmark page.
NVIDIA GeForce RTX 5090
*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
| Spec | Radeon RX 9070 XT | NVIDIA GeForce RTX 5090 |
|---|---|---|
| vram gb | 16 | 32 |
| vram type | GDDR6 | GDDR7 |
| tdp w | 304 | 575 |
| cuda cores | — | 21760 |
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, TechPowerUp, 3DMark, Overclocking.com, Tom's Hardware and PassMark; 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.
| Benchmark | Radeon RX 9070 XT | NVIDIA GeForce RTX 5090 |
|---|---|---|
| 3DMark Fire Strike (higher is better) Source: LANOC |
67,167 pts | — |
| 3DMark Fire Strike Extreme (higher is better) Source: LANOC |
33,887 pts | — |
| 3DMark Port Royal (higher is better) Source: TechPowerUp, LANOC, 3DMark |
18,240 points | 41,238 points |
| 3DMark Speed Way (higher is better) Source: Overclocking.com, LANOC, 3DMark |
6,267 pts | 15,488 points |
| 3DMark Steel Nomad (higher is better) Source: LANOC, Overclocking.com, 3DMark |
7,025 points | 11,906 points |
| 3DMark Steel Nomad (4K) (higher is better) Source: Tom's Hardware |
— | 14,133 points |
| 3DMark Time Spy (higher is better) Source: TechPowerUp, LANOC, 3DMark |
29,716 pts | 39,365 pts |
| 3DMark Time Spy Extreme (higher is better) Source: LANOC |
14,367 pts | — |
| PassMark G2D Mark (higher is better) Source: PassMark |
1,318 pts | 1,412 pts |
| PassMark G3D Mark (higher is better) Source: PassMark |
26,908 pts | 38,935 pts |
| Tom's Hardware GPU Hierarchy (lower is better) Source: Tom's Hardware |
719 rank | 2 rank |
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 9070 XT
- qwen3:8b: 711.0 tok/s (LocalLLaMA)
- Llama 3.2 1.5B: 163.0 tok/s (CraftRigs)
- llama2:7b: 142.9 tok/s (llama.cpp GitHub)
NVIDIA GeForce RTX 5090
- Qwen2.5-Coder-7B-Instruct: 5841.0 tok/s (Runpod)
- Llama 3.1 8B FP8: 420.0 tok/s (Runpod)
- gpt-oss:20b: 298.2 tok/s (Hardware Corner)
Quick answers
Which is faster, the Radeon RX 9070 XT or the NVIDIA GeForce RTX 5090?
The NVIDIA GeForce RTX 5090 is ahead of the Radeon RX 9070 XT on all 5 shared benchmarks on file, by 7% to 147%, including 147% in 3DMark Speed Way (15,488 vs 6,267).
Which is better for running local LLMs?
On gpt-oss:20b, the NVIDIA GeForce RTX 5090 generates 298.2 tokens/s (per Hardware Corner) against 91.9 for the Radeon RX 9070 XT (per hirokuze (GitHub)), best community run on file for each.
What did the Radeon RX 9070 XT and the NVIDIA GeForce RTX 5090 cost at launch?
The Radeon RX 9070 XT launched at a $599 MSRP and the NVIDIA GeForce RTX 5090 at $1,999. 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 →- How to Build a Windows 98 Retro PC in 2026
- Best Budget Gaming PC Build 2026 — ~$1,000 Parts List
- Best 1440p Gaming GPUs in 2026
- The Complete Voodoo5 5500 AGP Driver Guide (2026 Edition)
- Best Retro Handhelds in 2026 — From $35 to $500
- RTX 4070 Super vs RX 7800 XT — Which to Buy in 2026
- Emulation Hardware in 2026: FPGA, Software, and Cart-Reader Ecosystems
More reviews from the SpecPicks archive
Browse all reviews →- Best GPU for Llama 70B Local Inference in 2026: RTX 3060 12GB Dual vs RTX 3090 vs Gorgon Halo
- GLM-5.2 vs DeepSeek V4 on a 12GB RTX 3060: Which Open-Weights Model Wins?
- Open WebUI on a Raspberry Pi 4, Inference on an RTX 3060: The Split-Brain Home LLM Stack
- Best GPU for Qwen3 8B in 2026
- RX 9070 XT vs RTX 5080 for a 9800X3D Build (2026)
- Intel Axes BigDL: What It Means for CPU and Arc LLM Inference
- ComfyUI on an RTX 3060 12GB: A 2026 Local Stable Diffusion Setup Guide
- Build a Wireless Fingerprint Door Lock on a Raspberry Pi Zero W
- RetroPie Handheld Build Guide 2026: Hardware & Setup
- Moonshot AI Targets $30B: Can You Run a Kimi-Class Open Model on a 12GB GPU?
- Best Microphones for Streaming and Podcasting (2026)
- Dual GPU Llama.cpp Speedup: What Actually Helps
- Best PC Gaming Upgrades Under $30 in 2026
- Best Budget GPU for Stable Diffusion: Why the RTX 3060 12GB Still Wins
- Best Retro Emulator Consoles 2026: AMD GPU Buying Guide
- How to run Qwen 3 14B on NVIDIA GeForce RTX 5070
- HyperX QuadCast 2 vs Blue Yeti for Twitch in 2026
- Qwen 3.6 35B-A3B on RTX 3060 12GB: Local LLM Throughput Guide (2026)
- Raspberry Pi 5 Local LLM Server: Best Models for 8GB RAM in 2026
- Benchmarking Open Models for Agentic Tool Use on an RTX 3060
- Raspberry Pi 4 8GB in 2026: Still the Right Pick for Retro Emulation and Home Servers?
- RTX 3060 12GB vs RTX 3060 Ti 8GB for Local LLMs: VRAM Beats Bandwidth Every Time
- Automated Cabinet Lighting With a Raspberry Pi: Build Guide
- Running a 3dfx Voodoo5 5500 PCI in a Modern Board in 2026
More buying guides from SpecPicks
Browse all buying guides →- Best GPUs for 4K Gaming in 2026
- Best AM5 Motherboards for 2026
- Best GPU for Running 27B-32B Local LLMs in 2026
- Best CPUs for Content Creators in 2026
- Best Graphics Cards for Gaming in 2026
- Best 4K Monitors for Content Creators in 2026
- Best CPUs for Gaming in 2026
- Best DDR5 RAM for Gaming PCs in 2026
- Best Tools for Building and Repairing Retro PCs in 2026
- Best Gaming Monitors for 2026
- Best External SSDs for Content Creators in 2026
- Best CPU Coolers for 2026
- Best NVMe SSDs for Gaming in 2026
- Best Gaming Mice for 2026
- Best 1440p 240Hz Gaming Monitors in 2026
- Best Controllers for PC Gaming in 2026
- Best GPUs for Running Local LLMs in 2026
- Best Mechanical Keyboards for Gaming in 2026
- Best NVMe External Enclosures for 2026
- Best Retro Gaming Consoles & Handhelds for 2026
- Best PC Cases for Building in 2026
Hardware benchmark data on SpecPicks
All benchmarks →- 3dfx Voodoo5 6000 — benchmarks & specs
- AMD Ryzen 7 7700G — benchmarks & specs
- AMD Ryzen 5 Pro 7535U — benchmarks & specs
- GeForce RTX 3050 6 GB — benchmarks & specs
- AMD Ryzen 9 9955HX — benchmarks & specs
- Intel Arc A380E — benchmarks & specs
- Ryzen 5 3500U — benchmarks & specs
- Radeon RX 6700 XT — benchmarks & specs
- AMD Ryzen 9 5900HS — benchmarks & specs
- AMD Ryzen 3 5300G — benchmarks & specs
- DDR5-6000 CL30 64GB (2x32) — benchmarks & specs
- Apple M3 Ultra — benchmarks & specs
- NVIDIA GeForce RTX 2060 SUPER — benchmarks & specs
- AMD Ryzen 5 PRO 7540U — benchmarks & specs
- Apple M4 Pro 14 Core — benchmarks & specs
- AMD Ryzen 7 5800 — benchmarks & specs
- AMD Ryzen 9 5900XT — benchmarks & specs
- Quadro RTX 6000 — benchmarks & specs
- AMD Ryzen 7 7435HS — benchmarks & specs
- Apple M1 Ultra 20 Core — benchmarks & specs
- Radeon RX 6600 — benchmarks & specs
- Quadro RTX 4000 — benchmarks & specs
- Radeon RX 6750 GRE 12GB — benchmarks & specs
- AMD Ryzen 7 7745HX — benchmarks & specs

