V100 SXM2 specs
NVIDIA · Legacy · first available Jun 2017 (estimated). Dense compute, memory, bandwidth and power from the vendor's documents; each precision stays separate.
All accelerators · Compare with H100 SXM
Specifications
| Dense BF16 · FP32 accumulate | UnsupportedFP16 only: 125 TFLOPS; no BF16 tensor format. |
|---|---|
| Dense FP8 · FP32 accumulate | Unsupported |
| Dense FP4 · FP32 accumulate | Unsupported |
| Memory | 32 GB · HBM |
| Memory bandwidth | 0.9 TB/s |
| Maximum chip power | 300 W |
| Interconnect | NVLink 2 · 300 GB/s |
| Process node | TSMC 12FFN |
| First availability | Jun 2017est — estimated / secondary-source figure |
| Availability | Legacy |
First tensor-core GPU; FP16 only (no BF16).
Dense tensor rates with FP32 accumulation, never sparse or FP16-accumulate headline figures. Memory is in decimal GB. The est marker flags a value its vendor-document review did not confirm.
Rent it
$0.23/GPU-hrCheapest on-demand · Verda · observed
4 providers: AWS, Lambda, Oracle OCI, Verda. Quotes checked through .
V100 rental prices from every provider →Value ratios
| $ per dense BF16 PFLOP-hour | No dense BF16 rate |
|---|---|
| Dense BF16 TFLOPS per watt | No dense BF16 rate |
| Bandwidth bytes per 1,000 BF16 FLOP | No dense BF16 rate |
| $ per dense FP8 PFLOP-hour | No dense FP8 rate |
| Dense FP8 TFLOPS per watt | No dense FP8 rate |
| Bandwidth bytes per 1,000 FP8 FLOP | No dense FP8 rate |
| $ per GB of memory per hour | $0.007 |
Prices divide the cheapest on-demand list price; theoretical ratios do not measure application speed.
Sources
Specifications reviewed . SXM2 column; FP16 tensor only (no BF16, FP8 or FP4).
Rental price: Verda published pricing ↗ (1-GPU instance 1V100.6V; Tesla V100 16GB; 6 vCPUs and 23 GB RAM per instance; published per instance-hour, which is per GPU-hour).