NVIDIA Tesla V100 16GB (SXM2)
Compare NVIDIA Tesla V100 16GB (SXM2) against another machine in the Arena
Reference: Nvidia Tesla on Wikipedia
- Class
- GPU
- Memory
- 16 GB HBM2
- Runs (Q4_K_M)
- —
- Bandwidth
- 897 GB/s
- TDP
- 300 W
- Released
- 2017-06-01
- Price (US)
- $250 used as of 2026-08
- Price (Canada)
- CA$367.23 used sourced, as of 2026-08
Sourcing and disambiguation notes
SXM2 module: it needs a server board with an SXM2 socket or a third-party SXM2-to-PCIe carrier, and cannot go straight into a desktop PCIe slot. That is the single most important buying caveat for homelab builders. No new stock exists. Used listings for the 16GB card, both PCIe and SXM2 bundled with an adapter, cluster around $195-300 and average $238-280 -- an approximate street price rather than a single verified sale. The Canadian figure is the same SXM2-plus-adapter basis as the US one.
Specification sources
- https://cputronic.com/en/gpu/nvidia-tesla-v100-pcie-16-gb — press, 2026-08-22
Measurements
Decode is token generation — the speed you feel while an answer streams. Prefill is prompt processing — the wait before it starts. Why bandwidth predicts decode speed.
No records yet for this hardware. Know of a published benchmark on it? Submit the link.
What verified, single-source and estimated mean, and the same rows with every filter and sort in the benchmarks explorer.
Speed over time
What it can run
Fit is arithmetic, not a measurement — how it is computed. Against 16 GB; models that fit are listed largest first.
No modeled quant fits in 16 GB.