NVIDIA
Tesla Referral V100 SXM2 GPU NVIDIA
Tesla Referral V100 16G SXM2 GPU, NVLink For AI Deep Learning & High Performance Computing Why Buy This ✓ Original
Tesla Referral V100 16GB SXM2 GPU, server pulled ✓ Built-in NVLink support for multi-GPU cluster setup ✓ Fully tested before shipment, stable operation ✓ Ideal for AI training, inference, HPC scientific computing Product Overview
This is original server pulled NVIDIA Tesla Referral V100 16GB SXM2 GPU card. It supports NVLink interconnect for multi-GPU parallel computing. Widely used in AI deep learning model training, artificial intelligence inference and high-performance scientific computing. Real photos of actual item. Requires compatible SXM2 server heatsink and server platform.
Key Features
- Model: NVIDIA Tesla Referral V100 SXM2, 16GB HBM2 memory
- Support NVLink, enables high-speed multi-GPU communication
- Original server pull GPU, fully function tested
- SXM2 form factor, must be used with matching server heatsink
- Perfect for deep learning, AI training, HPC, simulation workloads
- Avoid using NVIDIA cards as boot GPUs to prevent driver conflicts and driver drops. Ubuntu Linux is recommended.
Specifications Item Detail Model NVIDIA
Tesla Referral V100 SXM2 Memory 16GB HBM2 Interface SXM2 with NVLink Condition Used original server pull, fully tested Important Pre-Purchase Notes ⚠ Please read carefully before purchasing:
- ✅ This is SXM2 module, not PCIe card. Needs compatible SXM2 server heatsink.
- ✅ Supports NVLink bridge for multi-GPU cluster deployment.
- ✅ Not for ordinary desktop PC, only works in SXM2 supported server.
- ✅ Real photos of the actual GPU you will receive.
Item Condition Used original GPU, fully tested, 1 month warranty, support NVLink connection.
- Second-hand server pulled Tesla Referral V100 SXM2 16G GPU
- Full function tested via nvidia-smi before packing
- Supports NVLink for multi-GPU parallel computing
- May have minor cosmetic marks, no impact on performance
Package Includes
1 × NVIDIA Tesla Referral V100 16G SXM2 GPU
Heatsink, NVLink bridge, server board NOT included
Warranty & Returns Warranty 1-month functional warranty Returns Accepted only for verified hardware defects Pre-Shipment Test Full function test before dispatch Shipping
- Dispatched within 1–3 business days after payment cleared
- Tracking number provided for every order
- Shipped with anti-static packaging for GPU protection
FAQ
Q: Can I install this GPU on regular desktop PC?
A: No. It is SXM2 form factor, only compatible with SXM2 supported server platform and dedicated heatsink.
Q: Does it come with NVLink bridge?
A: GPU only. NVLink bridge and heatsink are not included.
Q: What workloads fit this V100 SXM2?
A: AI deep learning training, model inference, HPC computing and simulation tasks.
Questions about server compatibility? Message us before ordering — we reply fast.
