RTX A6000

Ampere NVIDIA RTX A Series Creator Workstation Desktop GDDR6 2020 NVIDIA

Key Metrics

Architecture: Ampere
Launch Date: 2020-10-05
CUDA Cores: 10,752
Memory: 48 GB GDDR6
Memory Bandwidth: 768 GB/s
TDP/TBP: 300 W
Interface: PCIe 4.0 x16

Save this GPU, then open another GPU page to compare them side by side.

GPU Specs

Identity
Entry Type	Workstation GPU
Official Name	NVIDIA RTX A6000
GPU Codename	GA102
Architecture	Ampere
Launch Date	2020-10-05
Market Segment Notes	Workstation, Creator in NVIDIA RTX A Series
Core Specs
GPU Die	GA102
Chip Variant / Package	BGA-3328
Process Node	Samsung 8 nm
Transistor Count	28.3 billion
Die Size	628 mm²
CUDA Cores	10752
SM Count	84
TMUs	336
ROPs	112
AI Units / Tensor Cores	336
RT Units / RT Cores	84
Base Clock	1,410 MHz
Boost Clock	1,800 MHz
Pixel Fillrate	201.6 GP/s
Texture Fillrate	604.8 GT/s
FP32 Throughput	38.71 TFLOPS
FP16 Throughput	38.71 TFLOPS
FP64 Throughput	0.6 TFLOPS
Memory
Memory Size	48 GB
Memory Type	GDDR6
Memory Bus Width	384-bit
Memory Clock	2,000 MHz
Memory Bandwidth	768 GB/s
L2 Cache	6 MB
L1 Cache	128 KB
Board and Display
TDP / TBP	300 W
Recommended PSU	700 W
Power Connectors	1x 8-pin CPU (EPS-12V)
Interface	PCIe 4.0 x16
Display Outputs	4x DisplayPort 1.4a
Slot Width	Dual-slot, 267 mm length
Card Length	267 mm
Board Partner Model	Reference or OEM specification
Reference Design	NVIDIA reference design
Software
CUDA Compute Capability	8.6
Shader Model	6.8
DirectX Support	DirectX 12.2
Vulkan Support	Vulkan 1.4
OpenGL Support	OpenGL 4.6
OpenCL Support	OpenCL 3.0

Partial entry. Core specifications may be present, but history, variant notes, known issues, and source review may be incomplete.

Launch Context

Launch context should document the Ampere professional visualization generation, the Quadro-to-RTX naming transition, and 48 GB workstation memory role from official NVIDIA and board-partner material.

Performance Context

Performance context should focus on workstation applications, rendering, AI, visualization, and large-scene workflows only when benchmark methodology is clear.

Used-Market Caution

Before buying used, confirm bracket height, display adapters, mini-DP or DMS breakout cables, fan condition, VRAM errors, driver branch support, OEM firmware quirks, and whether the card needs workstation-specific airflow or auxiliary power.

Sources

Sources reviewed: 2026-08-13

This entry is still being source-checked.

Source Status

Source review pending. Treat narrative notes as incomplete until listed sources are reviewed.

Pending source review topics:

why it mattered
launch context
performance context
competitive context
driver and platform notes
variant notes

Data confidence and source status

Partial entry. Core specifications may be present, but history, variant notes, known issues, and source review may be incomplete.

Verified: core specifications and historical notes have been reviewed against listed sources.
Partial: core specifications may be present, but history, variant notes, known issues, and source review may be incomplete.
Imported: spec-only imported entry pending editorial source review.
Needs source review: narrative notes are incomplete until reviewed.

Primary / manufacturer sources

NVIDIA, RTX A6000 datasheet PDFPrimary datasheet source for RTX A6000 and Ampere professional positioning.Open source
PNY, NVIDIA RTX A6000 product pageBoard-partner source for CUDA, RT, Tensor core counts and 48 GB memory.Open source
Lenovo Press, ThinkSystem NVIDIA RTX A6000 48GB PCIe Active GPUOEM compatibility and lifecycle context for workstation/server deployments.Open source
NVIDIA RTX A6000 datasheetOfficial memory, interface, board power, and connector specificationsOpen source
NVIDIA workstation power guidelinesOfficial total board power and auxiliary connector tableOpen source

Independent reference sources

TechPowerUp GPU SpecsModel-level specification reference for this GPU entryOpen source

Derived datasets and spec dumps

RightNow GPU DatabaseOpen data inventory used for bulk normalization and cross-checking Derived/spec-dump source. Treat as a reference dataset, not independent historical verification.Open source

Found an incorrect spec, missing source, or variant issue?

Submit a correction