900-2G193-0030-001 NVIDIA L4 24GB GDDR6 2040MHz 300GB/s GPU

Get extra volume discount for 900-2G193-0030-001 and save cash:

Quantity	Unit price
	€3,620.00
To 4	€3,550.00 €3,620.00 (1.93% saved)
To 9	€3,480.00 €3,620.00 (3.87% saved)
From 10	€3,410.00 €3,620.00 (5.8% saved)

Product line:

Tesla

See all AI/Deep Learning GPUs

See other NVIDIA products

€3,620.00

Out of stock

Happyware Article Number:

SE3G8463-654742

Manufacturer Number:

900-2G193-0030-001

Brand:

NVIDIA

Are you a reseller?

Do you need help or a free consultation regarding our Services?

Contact our experts - we will assist You in finding the right services for Your requirements.

Do you need help?
Your contact

+49 4105 770000

Contact form

Description Technical details Warranty & Technical Support Reviews Bulk Deals

900-2G193-0030-001

Low-Profile Single-Slot PCIe AI Accelerator for Dense Servers with the 900-2G193-0030-001

The 900-2G193-0030-001 is an NVIDIA Tesla L4 PCIe AI accelerator designed for inference and media pipelines in space-constrained, high-density server platforms. Based on the 5 nm Ada Lovelace architecture, it delivers Tensor Core acceleration and 7424 CUDA cores while maintaining a low-profile, single-slot mechanical envelope suited to multi-GPU and mixed I/O configurations.

For memory-bound inference, the 900-2G193-0030-001 integrates 24GB GDDR6 on a 192-bit interface, providing 300 GB/s bandwidth to sustain LLM serving, recommender execution, and vision inference at consistent latency. PCIe 4.0 x16 host connectivity enables efficient integration into mainstream CPU servers, while keeping lanes available for NVMe storage, SmartNICs, and high-speed networking. Its passive, fanless design aligns with front-to-back airflow chassis and supports predictable thermal behavior in standardized racks.

Why Choose the 900-2G193-0030-001?

The 900-2G193-0030-001 targets a practical density-to-performance ratio for enterprises scaling inference across many nodes, where slot economy and power/thermal planning are as critical as throughput. It enables repeatable deployments in 1U/2U servers, supports rapid expansion through PCIe-based fleet standardization, and helps consolidate AI inference and video processing on a single accelerator class without redesigning the platform.

NVIDIA engineering emphasizes driver maturity, ecosystem integration, and lifecycle stability for production data centres. The 900-2G193-0030-001 fits modern operational standards by pairing Ada Lovelace efficiency with PCIe 4.0 compatibility and proven enterprise software enablement for secure, manageable GPU deployments.

Key Applications:

LLM inference and serving
Recommender system inference
Computer vision inference
Video transcoding and stream processing
Edge inference in dense servers

At Happyware - Server Hardware you benefit from non-binding offers tailored precisely to your deployment scope for the 900-2G193-0030-001, and an extended warranty up to 6 years to align GPU lifecycle planning with enterprise refresh cycles.

Technical details of 900-2G193-0030-001

GPU Model: L4

Series: Tesla

Memory Size: 24GB

Stream/CUDA Processors: 7424

Memory Type: GDDR6

Memory Bandwidth: 300GB/s

Base Clock: 795MHz

Boost Clock: 2040MHz

FLOPS: 29.70 TFLOPS (FP16), 29.70 TFLOPS (FP32), 0.46 TFLOPS (FP64)

GPU slot width: Single Slot

Architecture: Ada Lovelace

Form Factor: Low-Profile

Memory Interface: 192-bit

Chipset Manufacturer: NVIDIA

Manufacturing Process: 5nm

How long is the Warranty and how you can get technical support for 900-2G193-0030-001

Warranty:	36 months (part)
RMA through customer:	yes
RMA Url:	https://enterprise-support.nvidia.com/s/rma-serial-check
Description:	Items must be returned first before a replacement is sent. No guaranteed delivery times. On-site service is not standard.