NVIDIA

NVIDIA 699-21001-0230-610 A100 80GB PCIe GPU accelerator

Name: NVIDIA 699-21001-0230-610 A100 80GB PCIe GPU accelerator
Brand: NVIDIA
SKU: 699-21001-0230-610
Price: 44148 USD
Availability: InStock

MPN: 699-21001-0230-610·By: NVIDIA

Part #:699-21001-0230-610NewIn stock

Ships from the US.

$44,148.00

Qty

NVIDIA 699-21001-0230-610 A100 80GB PCIe GPU accelerator-

$44,148.00

Order by 1:00 PM CST for overnight shipping

Buying in volume or need a PO?

Have a target price for 699-21001-0230-610? Call +1 (551) 421-5162 or request a bulk quote and our sales team will reply within one business day.

Have questions?

Talk to our customer service team.

Phone Email

Product overview

The NVIDIA A100 80GB PCIe GPU, SKU 699-21001-0230-610, is an application accelerator designed for demanding compute environments. As an A100 80GB HBM2e PCIe 4.0 x16 GPU accelerator, it delivers high bandwidth connectivity through the PCIe 4.0 x16 interface for efficient data exchange with host systems. The 80GB HBM2e memory configuration is suited to workloads that require large memory capacity and high memory throughput. Classified as an expansion module application accelerator, this GPU is intended to be integrated into compatible server or workstation platforms via a PCIe slot. The product category highlights its role as a dedicated accelerator rather than a general graphics adapter. While detailed performance, networking ports and interfaces, and power consumption values are not specified, the defined model, memory type, memory capacity, and interface give IT procurement teams core information needed for initial solution planning and compatibility checks.

80GB HBM2e on board GPU memory for data intensive tasks
PCIe 4.0 x16 host interface for accelerator connectivity
Designed as an expansion module application accelerator
NVIDIA A100 80GB PCIe GPU model identification
SKU 699-21001-0230-610 for precise procurement
HBM2e based GPU accelerator for high bandwidth memory

Expedited and overnight shipping options are available at checkout, priced by destination and service.
Purchase orders accepted from qualified institutions.
Refunds within 14 days and replacements within 30 days of delivery with an RMA (25% restocking fee on customer-initiated returns), and every part carries our Standard 30-Day Warranty - see our return policy or contact US IT Hardware.

Featuring

Manufacturer part #	699-21001-0230-610
Brand	NVIDIA
Condition	New
Model	A100 80GB HBM2 PCI-E GPU Tensor Ampere Computing Accelerator Card
CUDA Cores	6912 CUDA cores
Form Factor	PCI-E GPU accelerator card
Memory Type	HBM2
Cooling Type	Passive cooling design
Product Line	NVIDIA A100
Product Type	GPU Tensor Ampere computing accelerator card
Tensor Cores	432 third-generation Tensor Cores
Product Class	Application Accelerator
Base GPU Clock	765 MHz
NVLink Support	Supports NVIDIA NVLink with three NVLink bridge connectors
Boost GPU Clock	1515 MHz
Memory Capacity	80 GB
FP16 Performance	312 TFLOPS FP16 or 624 TFLOPS FP16 with sparsity
FP32 Performance	19.5 TFLOPS FP32
FP64 Performance	9.7 TFLOPS (double-precision FP64)
GPU Architecture	Ampere
INT4 Performance	1248 TOPS INT4 or 2496 TOPS INT4 with sparsity
INT8 Performance	624 TOPS INT8 or 1248 TOPS INT8 with sparsity
Memory Bus Width	5120-bit
Total NVLink Lanes	96 total NVLink Rx and Tx lanes supported
BFLOAT16 Performance	312 TFLOPS BFLOAT16 or 624 TFLOPS BFLOAT16 with sparsity
NVLink Lane Data Rate	50 Gbps per NVLink lane (each direction)
PCI Express Generation	PCI Express Gen 4
NVLink Bridges Required	Uses three NVIDIA NVLink bridges for multi-GPU connectivity
FP64 Tensor Core Performance	19.5 TFLOPS FP64 Tensor Core
Multi-Instance GPU (MIG) Support	Up to 7 GPU instances with MIG
Tensor Float 32 (TF32) Performance	156 TFLOPS TF32 or 312 TFLOPS TF32 with sparsity

Performance

Peak Memory Bandwidth	Up to 1.94 TB/s
Maximum NVLink Bandwidth	600 GB/s total maximum NVLink bandwidth

Power Consumption

Thermal Design Power (TDP)	250 W GPU power

Networking Ports & Interfaces

Interface	PCI-E
Auxiliary Power Connector Type	One CPU 8-pin auxiliary power connector

Best for

Application acceleration
Suitable where a dedicated A100 80GB HBM2e PCIe 4.0 x16 GPU accelerator is required to offload and accelerate compute intensive tasks from the host system.
Expansion module deployments
Ideal for systems that accept expansion modules, using the A100 80GB PCIe GPU as an add in accelerator to extend compute capabilities.
PCIe 4.0 x16 platforms
Fits environments built around PCIe 4.0 x16 slots, allowing integration of the A100 80GB HBM2e GPU accelerator into compatible servers or workstations.
High memory capacity workloads
Applicable to workloads that benefit from 80GB of HBM2e memory on the GPU, where large datasets or models must be resident on accelerator memory.
HBM2e based accelerator evaluations
Useful for teams standardizing on HBM2e based GPU accelerators and needing an A100 80GB PCIe option identified by SKU 699-21001-0230-610.

Frequently asked questions

What is the memory configuration of the NVIDIA A100 80GB PCIe GPU 699-21001-0230-610?

This product is specified as an A100 80GB HBM2e PCIe 4.0 x16 GPU accelerator, indicating that it provides 80GB of on board HBM2e memory. The 80GB capacity and HBM2e memory type are defined in the category description, giving a clear indication of the accelerator’s memory resources for planning workloads.

Which interface does the NVIDIA A100 80GB PCIe GPU use for host connectivity?

According to the product category, this A100 80GB HBM2e GPU uses a PCIe 4.0 x16 interface. This means it is intended to be installed in a compatible PCI Express 4.0 x16 slot in the host system, functioning as an expansion module application accelerator for compute intensive tasks.

How should this NVIDIA A100 80GB PCIe GPU be integrated into a system?

The category identifies the device as an expansion module application accelerator, specifically an A100 80GB HBM2e PCIe 4.0 x16 GPU accelerator. It is designed to be added to a server or workstation via a PCIe 4.0 x16 slot, where it operates alongside the host CPU to accelerate workloads.

What key identifiers should procurement teams use for this A100 GPU?

Procurement teams should reference the brand NVIDIA, the model A100 80GB PCIe GPU, and the SKU 699-21001-0230-610. The category further specifies it as an A100 80GB HBM2e PCIe 4.0 x16 GPU accelerator, which helps distinguish it from other A100 variants and ensures accurate ordering and asset tracking.

Is this NVIDIA A100 solution positioned as a general graphics card or as a compute accelerator?

The product is categorized as an expansion module application accelerator and described as an A100 80GB HBM2e PCIe 4.0 x16 GPU accelerator. This classification indicates that it is intended primarily as a compute accelerator for integration into host systems, rather than as a general purpose graphics adapter.

Details

Shipping, warranty & returns

Customer reviews

No reviews yet — be the first to review this part.

Free ground shipping over $100

Secure SSL checkout. Visa, Mastercard, Amex and Discover accepted. ACH / wire available for businesses.

Buy with confidence

New, certified-refreshed and OEM stock with guaranteed compatibility, backed by warranty on every part.

Business account benefits

Net payment terms
Dedicated account rep
Warehousing & scheduled releases
Stock testing reports

Set up your business account →

Govt, federal & edu procurement

We process quotes, POs, stock documentation and compliance invoicing.

[email protected]

Manufacturer part #

699-21001-0230-610

Brand

NVIDIA

Condition

New

Model

A100 80GB HBM2 PCI-E GPU Tensor Ampere Computing Accelerator Card

CUDA Cores

6912 CUDA cores

Form Factor

PCI-E GPU accelerator card

Memory Type

HBM2

Cooling Type

Passive cooling design

Product Line

NVIDIA A100

Product Type

GPU Tensor Ampere computing accelerator card

Tensor Cores

432 third-generation Tensor Cores

Product Class

Application Accelerator

Base GPU Clock

765 MHz

NVLink Support

Supports NVIDIA NVLink with three NVLink bridge connectors

Boost GPU Clock

1515 MHz

Memory Capacity

80 GB

FP16 Performance

312 TFLOPS FP16 or 624 TFLOPS FP16 with sparsity

FP32 Performance

19.5 TFLOPS FP32

FP64 Performance

9.7 TFLOPS (double-precision FP64)

GPU Architecture

Ampere

INT4 Performance

1248 TOPS INT4 or 2496 TOPS INT4 with sparsity

INT8 Performance

624 TOPS INT8 or 1248 TOPS INT8 with sparsity

Memory Bus Width

5120-bit

Total NVLink Lanes

96 total NVLink Rx and Tx lanes supported

BFLOAT16 Performance

312 TFLOPS BFLOAT16 or 624 TFLOPS BFLOAT16 with sparsity

NVLink Lane Data Rate

50 Gbps per NVLink lane (each direction)

PCI Express Generation

PCI Express Gen 4

NVLink Bridges Required

Uses three NVIDIA NVLink bridges for multi-GPU connectivity

FP64 Tensor Core Performance

19.5 TFLOPS FP64 Tensor Core

Multi-Instance GPU (MIG) Support

Up to 7 GPU instances with MIG

Tensor Float 32 (TF32) Performance

156 TFLOPS TF32 or 312 TFLOPS TF32 with sparsity

Peak Memory Bandwidth

Up to 1.94 TB/s

Maximum NVLink Bandwidth

600 GB/s total maximum NVLink bandwidth

Interface

PCI-E

Auxiliary Power Connector Type

One CPU 8-pin auxiliary power connector