AMD Radeon RX 6900 XT, AI & Machine Learning Benchmarks & Specs

16GB · AI Score 2.3/100 · anchored estimate vs 51 measured cards

2.3 AI Score Includes estimates

We have not run AMD Radeon RX 6900 XT on our bench. These figures are anchored estimates, interpolated per workload against the 51 GPUs we did measure (confidence: low). On Llama 3.1 8B (Q4_K_M) AMD Radeon RX 6900 XT should deliver about 84.3 tokens/sec. Llama 3.3 70B does not fit. It needs roughly 42GB and this card has 16GB. 2 of the 12 workloads won't fit on 16GB at the tested precision, Qwen3 32B, Llama 3.3 70B. We publish those as hard gates rather than quietly dropping to a smaller quant. Image, video and 3D models aren't listed for this card. They're built for NVIDIA's CUDA, and on AMD they only run through ROCm workarounds whose speed depends more on the driver and OS than on the card, so we don't estimate them, and they count as not supported in the AI score. For text generation (llama.cpp runs natively on ROCm and Vulkan), the numbers above hold up.

AI & Machine Learning benchmark results

Text Generation tok/s 8

Qwen3 4B135.6

Llama 3.1 8B84.3

Qwen2.5-Coder 14B45.8

Workload	Result	Telemetry	Data
Qwen3 4B	135.6 tok/s	estimated	Est.
Llama 3.1 8B	84.3 tok/s	estimated	Est.
Qwen2.5-Coder 14B	45.8 tok/s	estimated	Est.
Gemma 4 26B A4B	✕ Won't fit needs ~20 GB	VRAM-gated at this precision	Est.
Qwen3 30B A3B	✕ Won't fit needs ~20 GB	VRAM-gated at this precision	Est.
Gemma 4 31B	✕ Won't fit needs ~22 GB	VRAM-gated at this precision	Est.
Qwen3 32B	✕ Won't fit	VRAM-gated at this precision	Est.
Llama 3.3 70B	✕ Won't fit	VRAM-gated at this precision	Est.

How this estimate is derived. This card hasn’t been through our bench yet, so its numbers are anchored estimates, interpolated from the 51 first-party measured cards (Bandwidth Theil-Sen ladder off NVIDIA anchors × vendor factor (LLM ×0.9, diffusion ×0.5); measured VRAM floors; recalibrated 2026-09-19 against published llama.cpp (CUDA/ROCm/Vulkan/SYCL) and Stable Diffusion results.). The VRAM “won’t fit” gates are exact, since they’re pure capacity limits. Confidence: low. Estimates are replaced with measured data as more silicon goes through the bench. Full methodology →

AMD Radeon RX 6900 XT specifications

Architecture	RDNA 2 (Navi 21)
Stream processors	5,120
VRAM	16GB GDDR6
Memory bus	256-bit
Memory bandwidth	512 GB/s
Boost clock	2,250 MHz
TDP	300 W
Process	7nm
Interface	PCIe 4.0 x16
Release date	2020-12-08
Launch MSRP	$999

Verdict, capable, but 16GB sets the ceiling

AMD Radeon RX 6900 XT scores 2.3/100, #71 of 102. It ran 3 of 12; 2 exceeded its 16GB. Figures are anchored estimates, not measurements, we flag that on every row.

Relative performance: where the AMD Radeon RX 6900 XT lands

100% = this card, AI & Machine Learning headline metric (AI Score). #32 of 61 desktop cards in this vertical.

GPU	%	AI Score
AMD Radeon RX 9070	104%	2.4
NVIDIA TITAN X (Pascal)	104%	2.4
AMD Radeon RX 6800 XT	100%	2.3
AMD Radeon RX 6800	100%	2.3
AMD Radeon RX 6900 XT	100%	2.3
AMD Radeon RX 6950 XT	100%	2.3
GeForce GTX 1080 Ti	100%	2.3
NVIDIA GeForce RTX 3080	100%	2.3
NVIDIA TITAN Xp	100%	2.3

One tier up: AMD Radeon RX 6800 (0%)
One tier down: AMD Radeon RX 6950 XT (0%)
Best value alternative ($179): Intel Arc B580 (-9%)

Same card, other workloads: AMD Radeon RX 6900 XT Gaming benchmarks

← All AI & Machine Learning GPU rankings

The silicon

Transistors	26,800 million
Die size	520 mm²
Process node	7 nm
Fabricated by	TSMC
Transistor density	51.5 million per mm²

Denser than 85% of the 746 cards we have silicon data for. Density is the clearest measure of what a process node bought: a card that gained it without growing the die got its speed from the fab rather than the architecture.

Silicon figures from Wikipedia (CC BY-SA 4.0). Benchmarks on this page are our own. Compare every chip.

What this card can build

Whole-job timings, composed from our measured per-model results on this card.

Workflow	Time	Energy	Basis
Full codebase review	21.8 min	n/a	estimate, 0 of 1 stage measured

Can't run: Short social clips (needs Qwen3 32B), 60-second AI short film (needs Qwen3 32B), 24-frame storyboard (needs Qwen3 32B), 6-panel comic page (needs Qwen3 32B), Long-form article batch (needs Llama 3.3 70B).