NVIDIA A800 80GB, AI & Machine Learning Benchmarks & Specs

80GB · AI Score 33.1/100 · anchored estimate vs 51 measured cards

33.1 AI Score Includes estimates

We have not run NVIDIA A800 80GB on our bench. These figures are anchored estimates, interpolated per workload against the 51 GPUs we did measure (confidence: high (sibling silicon)). On Llama 3.1 8B (Q4_K_M) NVIDIA A800 80GB should deliver about 162.6 tokens/sec. Stepping up to Qwen3 32B it should hold roughly 45.5 tok/s. The full Llama 3.3 70B still runs, at about 24.4 tok/s. For image generation, SDXL should run near 8.18 it/s, and FLUX.1-dev at 1.99 it/s. All 12 workloads fit in 80GB. There is no model in our suite this card has to turn down.

AI & Machine Learning benchmark results

Text Generation tok/s 5

Qwen3 4B197

Llama 3.1 8B162.6

Qwen2.5-Coder 14B89.5

Qwen3 32B45.5

Llama 3.3 70B24.4

Workload	Result	Telemetry	Data
Qwen3 4B	197 tok/s	estimated	Est.
Llama 3.1 8B	162.6 tok/s	estimated	Est.
Qwen2.5-Coder 14B	89.5 tok/s	estimated	Est.
Qwen3 32B	45.5 tok/s	estimated	Est.
Llama 3.3 70B	24.4 tok/s	estimated	Est.

Image Generation images/min 3

Stable Diffusion XL16.36

Z-Image Turbo9.975

FLUX.1 dev4.264

Workload	Result	Telemetry	Data
Stable Diffusion XL	16.36 images/min	estimated	Est.
FLUX.1 dev	4.26 images/min	estimated	Est.
Z-Image Turbo	9.98 images/min	estimated	Est.

Image Editing images/min 2

Workload	Result	Telemetry	Data
FLUX.1 Kontext dev	1.99 images/min	estimated	Est.
Qwen-Image-Edit	1.64 images/min	estimated	Est.

Video Generation frames/s 2

Workload	Result	Telemetry	Data
LTX-Video (distilled)	8.95 frames/s	estimated	Est.
Wan 2.2 5B (720p)	0.66 frames/s	estimated	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 (Sibling to measured A100 80GB SXM: identical compute+bandwidth (export SKU, NVLink cut to 400GB/s, no single-GPU impact)). The VRAM “won’t fit” gates are exact, since they’re pure capacity limits. Confidence: high (sibling silicon). Estimates are replaced with measured data as more silicon goes through the bench. Full methodology →

NVIDIA A800 80GB specifications

Architecture	Ampere
CUDA cores	6,912
VRAM	80GB HBM2e
Memory bus	5120-bit
Memory bandwidth	2039 GB/s
Boost clock	1,410 MHz
TDP	400 W
Process	TSMC 7nm
Interface	SXM4
Release date	2022-11-01
Launch MSRP	$15,000

Verdict, NVIDIA A800 80GB on real AI workloads

NVIDIA A800 80GB scores 33.1/100, #14 of 102. It ran all 12 workloads. Figures are anchored estimates, not measurements, we flag that on every row.

Relative performance: where the NVIDIA A800 80GB lands

100% = this card, AI & Machine Learning headline metric (AI Score). #12 of 21 datacenter cards in this vertical.

GPU	%	AI Score
NVIDIA H100 NVL	176%	58.1
NVIDIA RTX PRO 6000 Blackwell Server Edition	151%	49.9
NVIDIA H100 PCIe	150%	49.8
NVIDIA A100 80GB SXM4	100%	33.1
NVIDIA A800 80GB	100%	33.1
NVIDIA A100 80GB PCIe	96%	31.7
NVIDIA L40S	84%	27.7
NVIDIA L40	59%	19.5
NVIDIA A40	53%	17.7

One tier up: NVIDIA A100 80GB SXM4 (0%)
One tier down: NVIDIA A100 80GB PCIe (-4%)

← All AI & Machine Learning GPU rankings

The silicon

Transistors	54,200 million
Die size	826 mm²
Process node	7 nm
Fabricated by	TSMC
Transistor density	65.6 million per mm²

Denser than 88% 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
24-frame storyboard	2.9 min	n/a	estimate, 0 of 2 stages measured
60-second AI short film	4 min	n/a	estimate, 0 of 3 stages measured
6-panel comic page	4.7 min	n/a	estimate, 0 of 3 stages measured
Character sheet, 12 poses	6.3 min	n/a	estimate, 0 of 2 stages measured
Full codebase review	11.2 min	n/a	estimate, 0 of 1 stage measured
Short social clips	13.6 min	n/a	estimate, 0 of 3 stages measured
Long-form article batch	19.1 min	n/a	estimate, 0 of 1 stage measured
Product photo shoot	22.5 min	n/a	estimate, 0 of 2 stages measured
Photo restoration batch	50.2 min	n/a	estimate, 0 of 1 stage measured

Rental price

Renting this card costs $0.601/hour right now. We started tracking on 2026-10-04, so there is not enough history to chart yet. The line fills in as prices move.