Key takeaways
- 10 listings match this collection right now.
- The SXM2 version of the V100 mounts to a proprietary NVLink baseboard rather than a standard PCIe slot, so it only fits servers built for that specific mezzanine form factor.
- The PCIe V100 draws up to 250 watts through an 8-pin CPU-style connector, and the host chassis needs both that connector and airflow to cool a passively cooled card.
- Volta needs above 4G decoding and large BAR support in host firmware, and its CUDA compute capability trails Ampere and Hopper for frameworks that need the latest instruction set.
The Tesla V100 is the Volta-generation datacenter accelerator with 16GB or 32GB of HBM2 and tensor cores, in a PCIe card or an SXM2 module for NVLink systems. It remains the most cost-effective route to HBM-class bandwidth for training and HPC workloads that do not need Ampere or Hopper features.
10 listings in this collection, filtered from Graphics Cards. The part number and manufacturer are stated on every listing, and the condition wherever the catalogue records one. For quantity, request a volume quote.
How to choose gpus and accelerators
Datacenter cards (Tesla, A-series) are passively cooled and expect server airflow; workstation cards (Quadro, RTX A) carry their own fans and display outputs. Match the slot (PCIe generation and width), the auxiliary power the card needs, and the memory the workload needs: inference and video work fits in 16 GB to 24 GB, training and large models want 32 GB to 80 GB.
Other gpus and accelerators collections
Browse every collection by specification
Account
