Skip to content

300,000+ products in stockAI infrastructure + enterprise IT hardware. Genuine product, full manufacturer warranty.

699-2G506-0210-300 — NVIDIA A100 80GB HBM2e SXM4 GPU Module (Ampere architecture, NVLink, for HGX A100 baseboards)

MPN: 699-2G506-0210-300|By: NVIDIA

NVIDIA A100 80GB HBM2e SXM4 GPU Module (Ampere architecture, NVLink, for HGX A100 baseboards)

In Stockships worldwide

Condition:
New — authorized channel
Warranty:
Full manufacturer warranty

Get pricing and availability

Email + WhatsApp so we always reach you.

Prefer your own inbox? Email us about this part

Have questions?

Talk to a hardware specialist.

WhatsAppEmail

Bulk order? Request volume pricing

Product overview

NVIDIA lists 699-2G506-0210-300 as an A100 80GB HBM2e SXM4 GPU Module (ampere architecture, NVLink, for HGX A100 baseboards). Accelerators are limited by the chassis before they are limited by the silicon: slot width, auxiliary power and airflow decide what a server can actually take. Tell us the platform and we will confirm it fits.

Ask for the datasheet and firmware notes with the quote; we would rather you check our claim than take it.

Featuring

Capacity
80GB
Manufacturer
NVIDIA

Physical characteristics

Product Type
GPUs & Accelerators
  • Expedited shipping options are available — rates confirmed with your quote.
  • Purchase orders accepted from qualified institutions.
  • Compatibility is validated against your server platform before dispatch. Send us the model and we will confirm support.

Specifications for 699-2G506-0210-300 as published by NVIDIA. Where a value is absent it is because we could not verify it, not because it does not exist — ask and we will check.

Where the NVIDIA 699-2G506-0210-300 is deployed

  • Sourcing 699-2G506-0210-300 against a build sheet that names it explicitly
  • Adding accelerators to servers that are already racked
  • Replacing a failed accelerator in a production node

Who this part is for

  • Research groups sizing a training or inference node to a budget
  • Procurement buying accelerators on a purchase order at lead time
  • Integrators building to a specified accelerator and chassis pairing

What you get ordering this from us

Ordered by part number

The order is placed against 699-2G506-0210-300, not against a description of it, so there is no room for a substitution you did not agree to.

Volume pricing

The listed price is per unit; tiered pricing applies at quantity and purchase orders are accepted.

Answered by a human

A quote comes back with the reasoning attached, including the cases where we think you should buy something else.

Questions buyers ask about this part

Is 699-2G506-0210-300 genuine and is the warranty intact?
Yes — new and sealed, with the NVIDIA warranty and the firmware update path intact. We do not sell refurbished stock as new, and we do not source grey-market units.
Will 699-2G506-0210-300 work in my server?
Send the chassis model. Slot width, auxiliary power connectors and airflow decide this before the silicon does, and a passively cooled card in the wrong chassis will throttle or shut down.
Do you accept purchase orders and quote at volume?
Yes — POs from qualified organisations, net terms on approval, and banded pricing once the quantity is worth banding. Send the number and we will price it properly.

Compare other GPUs & Accelerators we stock

Browse all GPUs & Accelerators or everything we stock from NVIDIA.

Specifying a system, not just the part?

Accelerators are usually bought as part of a node rather than on their own — the chassis, power and cooling have to match the card.