Radeon RX 6900 XT vs NVIDIA GeForce GTX 1660 — Specs & Performance Compared
Verdict: The Radeon RX 6900 XT is ahead of the NVIDIA GeForce GTX 1660 on all 4 shared benchmarks on file, by 37% to 288%, including 218% in 3DMark Time Spy (17,564 vs 5,521). On llama3.1:8b, the Radeon RX 6900 XT generates 42.4 tokens/s (per LocalScore (Mozilla Builders)) against 16.7 for the NVIDIA GeForce GTX 1660 (per DatabaseMart (GPU hosting blog)), best community run on file for each.
Side-by-side GPU comparison of the Radeon RX 6900 XT and the NVIDIA GeForce GTX 1660. Released 2020 (Radeon RX 6900 XT), 2019 (NVIDIA GeForce GTX 1660). MSRP range: $999 · $219. Specs, synthetic benchmarks, AI inference tok/s, and Amazon availability below; each column header links to that part's benchmark page.
NVIDIA GeForce GTX 1660
*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 6900 XT | NVIDIA GeForce GTX 1660 |
|---|---|---|
| vram gb | 16 | 6 |
| vram type | — | GDDR5 |
| tdp w | — | 120 |
| cuda cores | — | 1408 |
Synthetic benchmarks
Each row says whether a higher or a lower number is better (ranks and times are lower-is-better). Scores come from Tom's Hardware, 3DMark, Notebookcheck, HWBOT, CpuTronic, NanoReview, TopCPU.net, Legit Reviews, GPU-Monkey 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 6900 XT | NVIDIA GeForce GTX 1660 |
|---|---|---|
| 3DMark Fire Strike (higher is better) Source: Tom's Hardware, 3DMark, Notebookcheck |
40,894 pts | 12,365 points |
| 3DMark Fire Strike Ultra (higher is better) Source: Notebookcheck, 3DMark |
15,204 points | — |
| 3DMark Port Royal (higher is better) Source: HWBOT, Notebookcheck, 3DMark |
11,583 points | — |
| 3DMark Speed Way (higher is better) Source: HWBOT, 3DMark |
4,389 points | — |
| 3DMark Steel Nomad (higher is better) Source: 3DMark, CpuTronic |
4,060 points | 1,047 points |
| 3DMark Steel Nomad Light (higher is better) Source: NanoReview |
20,093 points | — |
| 3DMark Time Spy (higher is better) Source: Tom's Hardware, TopCPU.net, Legit Reviews, Notebookcheck, CpuTronic, 3DMark |
17,564 pts | 5,521 points |
| 3DMark Time Spy (Graphics score) (higher is better) Source: Legit Reviews |
18,486 points | — |
| 3DMark Time Spy (Overall score) (higher is better) Source: Legit Reviews |
17,564 points | — |
| 3DMark Time Spy Graphics (higher is better) Source: GPU-Monkey |
— | 5,395 points |
| 3DMark Time Spy Graphics Score (higher is better) Source: Legit Reviews |
18,486 points | — |
| PassMark G2D Mark (higher is better) Source: PassMark |
1,062 pts | 776 pts |
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 6900 XT
- qwen3:8b: 711.0 tok/s (LocalLLaMA)
- llama2:7b: 108.0 tok/s (llama.cpp GitHub perf thread (Vulkan scoreboard))
- llama2:7b: 108.0 tok/s (llama.cpp GitHub Discussion #10879 (Vulkan))
NVIDIA GeForce GTX 1660
- qwen3:0.6b: 47.1 tok/s (LocalLLaMA)
- llama2:7b: 41.4 tok/s (llama.cpp GitHub (discussion #15013))
- llama2:7b: 41.4 tok/s (llama.cpp GitHub CUDA Performance Discussion)
Quick answers
Which is faster, the Radeon RX 6900 XT or the NVIDIA GeForce GTX 1660?
The Radeon RX 6900 XT is ahead of the NVIDIA GeForce GTX 1660 on all 4 shared benchmarks on file, by 37% to 288%, including 218% in 3DMark Time Spy (17,564 vs 5,521).
Which is better for running local LLMs?
On llama3.1:8b, the Radeon RX 6900 XT generates 42.4 tokens/s (per LocalScore (Mozilla Builders)) against 16.7 for the NVIDIA GeForce GTX 1660 (per DatabaseMart (GPU hosting blog)), best community run on file for each.
What did the Radeon RX 6900 XT and the NVIDIA GeForce GTX 1660 cost at launch?
The Radeon RX 6900 XT launched at a $999 MSRP and the NVIDIA GeForce GTX 1660 at $219. 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 →- RTX 4070 Super vs RX 7800 XT — Which to Buy in 2026
- Emulation Hardware in 2026: FPGA, Software, and Cart-Reader Ecosystems
- How to Build a Windows 98 Retro PC in 2026
- The Complete Voodoo5 5500 AGP Driver Guide (2026 Edition)
- Best Retro Handhelds in 2026 — From $35 to $500
- Best 1440p Gaming GPUs in 2026
- Best Budget Gaming PC Build 2026 — ~$1,000 Parts List
More reviews from the SpecPicks archive
Browse all reviews →- Intel Arc AI Software: oneAPI, OpenVINO & LLM Guide
- Best SATA SSD for the PS4 Pro in 2026 (Cheap Speed Upgrade)
- Can the RTX 3060 12GB Still Game at 1440p in 2026?
- Best Budget Local-AI Workstation Parts in 2026: 5 Picks
- Raspberry Pi Projects on GitHub: 2025 Guide
- Qwen 3.6 35B-A3B on RTX 3060 12GB: Local LLM Throughput Guide (2026)
- Crucial BX500 vs Samsung 970 EVO Plus: SATA vs NVMe
- Best Budget GPU for Stable Diffusion and SDXL in 2026
- Amstrad CPC Emulator Now Runs on the Raspberry Pi Pico 2
- Quiet RTX 3060 12GB Local LLM Box: Build Notes from a Real Setup
- Build a Self-Hosted AI Home Lab on a Raspberry Pi 4 8GB in 2026
- vLLM vs llama.cpp for Single-User Chat on a 12GB GPU (2026)
- Best GPU for Local LLMs Under $400 in 2026
- HyperX QuadCast 2 vs Blue Yeti for Streaming in an Untreated Room
- Raspberry Pi 4 8GB vs Pi Zero W for RetroPie: Which Board Wins?
- Zero-CPU-Overhead OV7725 Camera Streaming on RP2350
- RTX 4090 AI TOPS in 2026: What the Number Means
- Why Nvidia Killed 3dfx — and How Voodoo Still Shapes Every GPU Launch
- 42 LLMs Tested for Apocalypse Compliance: What Data Shows
- Local LLM PC Build 2026: What 12GB of VRAM Really Runs
- Steam Deck OLED vs ROG Xbox Ally: 2026 Comparison
- ComfyUI on an RTX 3060 12GB: Local Stable Diffusion Setup and Real Throughput
- GPU-Accelerated Autorouter Handles Monstrous PCB Designs
- Local PDF-to-Audiobook on a Raspberry Pi 5: Kokoro 82M + Qwen + llama.cpp
More buying guides from SpecPicks
Browse all buying guides →- Best External SSDs for Content Creators in 2026
- Best NVMe External Enclosures for 2026
- Best CPU Coolers for 2026
- Best GPUs for Running Local LLMs in 2026
- Best Gaming Monitors for 2026
- Best GPU for Running 27B-32B Local LLMs in 2026
- Best Gaming Mice for 2026
- Best 1440p 240Hz Gaming Monitors in 2026
- Best Retro Gaming Consoles & Handhelds for 2026
- Best PC Cases for Building in 2026
- Best Mechanical Keyboards for Gaming in 2026
- Best Controllers for PC Gaming in 2026
- Best CPUs for Gaming in 2026
- Best CPUs for Content Creators in 2026
- Best AM5 Motherboards for 2026
- Best DDR5 RAM for Gaming PCs in 2026
- Best Graphics Cards for Gaming in 2026
- Best GPUs for 4K Gaming in 2026
- Best Tools for Building and Repairing Retro PCs in 2026
- Best NVMe SSDs for Gaming in 2026
- Best 4K Monitors for Content Creators in 2026
Hardware benchmark data on SpecPicks
All benchmarks →- AMD Ryzen 3 PRO 5355G — benchmarks & specs
- Ryzen 3 3200G — benchmarks & specs
- AMD Ryzen 9 7945HX3D — benchmarks & specs
- AMD Ryzen 5 5600T — benchmarks & specs
- Apple M4 9 Core — benchmarks & specs
- GeForce RTX 5060 Ti 16GB — benchmarks & specs
- Apple M4 — benchmarks & specs
- AMD Ryzen 5 7600X3D — benchmarks & specs
- Quadro RTX 5000 (Mobile) — benchmarks & specs
- NVIDIA GeForce RTX 3060 Ti — benchmarks & specs
- AMD Ryzen 3 PRO 5350G — benchmarks & specs
- Ryzen 7 5700G — benchmarks & specs
- AMD Ryzen 7 9700F — benchmarks & specs
- NVIDIA H100 NVL 94GB — benchmarks & specs
- AMD Ryzen 5 5500H — benchmarks & specs
- Radeon RX 6500 XT — benchmarks & specs
- NVIDIA L40S 48GB — benchmarks & specs
- Apple M1 8 Core 3200 MHz — benchmarks & specs
- Radeon RX 5700 XT — benchmarks & specs
- AMD Ryzen 5 5600XT — benchmarks & specs
- AMD Ryzen 5 7535H — benchmarks & specs
- NVIDIA Jetson Orin Nano 8GB — benchmarks & specs
- Ryzen 9 9950X — benchmarks & specs
- GRID RTX6000P-2B — benchmarks & specs

