The PNY L4 Tensor professional graphics card is based on the NVIDIA Ada Lovelace architecture and is designed to accelerate video, artificial intelligence, visual computing, graphics, and virtualization in servers, data centers, and cloud environments. It features passive cooling and a low-profile design, and supports up to 1,040 simultaneous AV1 video streams at 720p30. It includes 4th-generation Tensor Cores and 3rd-generation RT Cores, as well as support for NVIDIA DLSS 3, NVIDIA Omniverse, and NVIDIA RTX vWS.
- Processor manufacturer: Nvidia
- Type: Professional graphics card
- Interface: PCI Express x16 4.0
- Memory: 24 GB GDDR6
- Memory speed: 12.5 Gbps
- Memory bus: 192-bit
- Maximum resolution: 7680 × 4320 pixels
- Power connector: 8-pin
- Cooling: Passive
- Design: Low profile
- Power consumption: 72 W
- Video support: Up to 1,040 simultaneous AV1 streams at 720p30
- Tensor Cores: 4th generation
- RT Cores: 3rd generation
- Supported technologies: NVIDIA DLSS 3, NVIDIA Omniverse, NVIDIA RTX vWS
- NVIDIA AI Enterprise support: Available as an additional license
- Color: Black
- Product Number
- TCSL4PCIE-PB
Main Specifications
- Processor Manufacturer
- Nvidia
- Chipset Model
- L4 Tensor
- Interface
- PCI Express x16 4.0
- Maximum Resolution
- 7680x4320 pixels
- Parallel Processing Technology
- Not Supported
- Type
- Professional
- Cuda Cores
- -
- Device Group
- -
- Colour
- Black
Special Features
- RGB Lighting
- No
- OverClock (OC)
- No
- Air-cooled
- No
- Low Profile
- Yes
- Passive Cooling
- Yes
- Water-cooled
- No
- External Graphics Card (eGPU)
- No
- LHR
- No
Memory
- Memory Capacity
- 24 GB
- Memory type
- GDDR6
- Memory Speed
- 12.5 Gbps
- Memory bus
- 192 bit
Connectivity
- Dual Link DVI
- No
System Requirements
- Minimum Power Supply
- -
- Power Connectors
- 8-pin
Top Specs
- Memory
- 24GB, GDDR6
- Connectivity
- -
Important information
Specifications are collected from official manufacturer websites. Please verify the specifications before proceeding with your final purchase. If you notice any problem you can report it here.