NVIDIA Tesla P4 GPU 8GB 900-2G414-0000-000 DDR5 Pascal CUDA PCIe x16 for Deep Learning, AI, HPC, Analytics and Research

(8 customer reviews)


Out of stock

This product has been discontinued/EOL. Please consider the new Tesla T4 GPU from NVIDIA.

Ideal for your Advanced Digital Transformation Applications : Video Processing, Big Data, Hyperconverged Appliances, Internet of Things (IoT), In-Memory Analytics, Machine Learning (ML), Artificial Intelligence (AI) and intensive Data Center or Hyperscale Infrastructure Applications. The NVIDIA Tesla GPUs are very suitable for autonomous cars, molecular dynamics, computational biology, fluid simulation etc and even for advanced Virtual Desktop Infrastructure (VDI) applications.

In the new era of AI and intelligent machines, deep learning is shaping our world like no other computing model in history. Interactive speech, visual search, and video recommendations are a few of many AI-based services that we use every day. Accuracy and responsiveness are key to user adoption for these services. As deep learning models increase in accuracy and complexity, CPUs are no longer capable of delivering a responsive user experience. The NVIDIA Tesla P4 is powered by the revolutionary NVIDIA Pascal™ architecture and purpose-built to boost efficiency for scale-out servers running deep learning workloads, enabling smart responsive AI-based services. It slashes inference latency by 15X in any hyperscale infrastructure and provides an incredible 60X better energy efficiency than CPUs. This unlocks a new wave of AI services previous impossible due to latency limitations.

As a NVIDIA Preferred Solution Provider, we are authorized by the manufacturer and proudly deliver only original factory packaged products. 

Key Features

  • Manufacturer Part Number: 900-2G414-0000-000
  • Sold and supported by NVIDIA
  • Small form-factor, 50/75-Watt design fits any scaleout server
  • Passively cooled board
  • 8 GB GDDR5 memory
  • INT8 operations slash latency by 15X.
  • Delivers 21 TOPs (TeraOperations per second) of inference performance
  • Hardware-decode engine capable of transcoding and inferencing 35 HD video streams in real time.
  • Low Profile PCI-e bracket (for Full-height bracket please select option below for 900-2G414-0000-001)
  • 3 Years Manufacturer’s Warranty


The NVIDIA® Tesla® P4 is a single-slot, low profile, 6.6 inch PCI Express Gen3 GPU Accelerator with an NVIDIA® Pascal™ graphics processing unit (GPU). The Tesla P4 has 8 GB GDDR5 memory and a 75 W maximum power limit. The Tesla P4 is offered as a 75 W or 50 W passively cooled board that requires system air flow to properly operate the card within thermal limits. The NVIDIA Tesla P4 features optimized INT8 instructions aimed at deep learning inference computations. As a result, the NVIDIA Tesla P4 delivers 21 TOPs (TeraOperations per second) of inference performance, enabling smart responsive artificial intelligence (AI)-based services. For performance optimization this board utilizes NVIDIA GPU Boost™, which will dynamically adjust the GPU clock to maximize performance within thermal limits.

Responsive Experience with Real-Time Inference

Responsiveness is key to user engagement for services such as interactive speech, visual search, Internet of Things (IoT) and video recommendations. As models increase in accuracy and complexity, CPUs are no longer capable of delivering a responsive user experience. The Tesla P4 delivers 22 TOPs of inference performance with INT8 operations

100X Higher Throughput to Keep Up with Expanding Data

50x Higher Throughput to Keep Up with Expanding Workloads

The volume of data generated every day in the form of sensor logs, images, videos, and records is economically impractical to process on CPUs. Volta-powered Tesla V100 GPUs give data centers a dramatic boost in throughput for deep learning workloads to extract intelligence from this tsunami of data. A server with single Tesla V100 can replace up to 50 CPU-only servers for deep learning inference workloads, so you get dramatically higher throughput with lower acquisition cost.

A Dedicated Decode Engine for New AI-based Video Services

A Dedicated Decode Engine for New AI-based Video Services

The Tesla P4 GPU can analyze up to 39 HD video streams in real time, powered by a dedicated hardware-accelerated decode engine that works in parallel with the NVIDIA® CUDA® cores performing inference. By integrating deep learning into the video pipeline, customers can offer new levels of smart, innovative video services that facilitate video search and other video-related services.

Unprecedented Efficiency for Low-Power Scale-out Servers

Unprecedented Efficiency for Low-Power Scale-out Servers

The ultra-efficient Tesla P4 GPU accelerates density-optimized scale-out servers with a small form factor and
50/75 W power footprint design. It delivers an incredible 52X better energy efficiency than CPUs for deep learning inference workloads so that hyperscale customers can scale within their existing infrastructure and service the exponential growth in demand for AI-based applications.

Faster Deployment With NVIDIA TensorRT™ and DeepStream SDK

Faster Deployment With NVIDIA TensorRT™ and DeepStream SDK

NVIDIA TensorRT is a high-performance neural network inference engine for production deployment of deep learning applications. It includes libraries to streamline deep learning models for production deployment, taking trained neural nets—usually in 32-bit or 16-bit data—and optimizing them for reduced-precision INT8 operations on Tesla P4, or FP16 on Tesla V100. NVIDIA DeepStream SDK taps into the power of Tesla GPUs to simultaneously decode and analyze video streams.

CUDA Ready

CUDA® is a parallel computing platform and programming model developed by NVIDIA for general computing on graphical processing units (GPUs). With CUDA, developers are able to dramatically speed up computing applications by harnessing the power of GPUs.

In GPU-accelerated applications, the sequential part of the workload runs on the CPU – which is optimized for single-threaded performance – while the compute intensive portion of the application runs on thousands of GPU cores in parallel. When using CUDA, developers program in popular languages such as C, C++, Fortran, Python and MATLAB and express parallelism through extensions in the form of a few basic keywords.

The CUDA Toolkit from NVIDIA provides everything you need to develop GPU-accelerated applications. The CUDA Toolkit includes GPU-accelerated libraries, a compiler, development tools and the CUDA runtime.

Performance Specifications for NVIDIA Tesla P4, P40 and V100 Accelerators

Tesla V100: The Universal Datacenter GPU Tesla P4 for Ultra-Efficient Scale-Out Servers Tesla P40 for Inference Throughput Servers
Single-Precision Performance (FP32) 14 teraflops (PCIe)
15.7 teraflops (SXM2)
5.5 teraflops 12 teraflops
Half-Precision Performance (FP16) 112 teraflops (PCIe)
125 teraflops (SXM2)
Integer Operations (INT8) 22 TOPS* 47 TOPS*
GPU Memory 16 GB HBM2 8 GB 24 GB
Memory Bandwidth 900 GB/s 192 GB/s 346 GB/s
System Interface/Form Factor Dual-Slot, Full-Height PCI Express Form Factor
SXM2 / NVLink
Low-Profile PCI Express Form Factor Dual-Slot, Full-Height PCI Express Form Factor
Power 250W (PCIe)
300W (SXM2)
50 W/75 W 250 W
Hardware-Accelerated Video Engine 1x Decode Engine, 2x Encode Engines 1x Decode Engine, 2x Encode Engines

*Tera-Operations per Second with Boost Clock Enabled

Additional information

Weight 8 lbs

8 reviews for NVIDIA Tesla P4 GPU 8GB 900-2G414-0000-000 DDR5 Pascal CUDA PCIe x16 for Deep Learning, AI, HPC, Analytics and Research

  1. Ashish

    We ordered NVIDIA Tesla P4 GPU 8GB DDR5 and got immediate reply from Dihuni. Prompt shipping & updates on delivery. Highly recommended for any order.

  2. M.P

    We got this in less than a week from Dihuni. Very fast service with regular updates, tracking information and great price on the gpu.

  3. Arthur

    we are using this nvidia product for deep learning and it works great.

  4. Tony

    the price point of the P4 GPU makes it very affordable and delivers the performance we need.

  5. L.Y

    We are happy to find an authorized NVIDIA partner that delivers. I don’t recommend buying such high end product from online stores that sell this product without proper authorization from manufacturer. Dihuni kept us posted about our order and although our package shipped out a couple days after initial date, we were happy that we were properly informed. Nice job Dihuni!

  6. Jennifer

    We have ordered this product from Dihuni multiple times and each time they have come through with on-time delivery and top service!!

  7. D Chang

    We have installed these gpus in our Supermicro 4u servers and they are very good at reasonable price.

  8. Shankar

    We ordered a couple of these from you. Your prices are better than competition. Keep up the nice work.

Add a review

Your email address will not be published. Required fields are marked *