NVIDIA · Volta · 2017
NVIDIA V100. Rental by the hour, on demand.
From $3.35/hr per GPU on demand, or prepay a 2-GPU machine at $3.19/GPU-hr for 1 month, or one GPU at $3.09/GPU-hr for 3 months.
V100 on demand
from$3.35/hr per GPU
- Availability
- Available now · 1 region
- Prepaid
- 1 mo $3.19/hr3 mo $3.09/hr6 mo $2.99/hr12 mo $2.85/hr
- Billing
- By the second on most GPUs, a few by the minute
- 8+ GPUs or a month
- Launch any size yourself, or prepay a machine for 1 to 12 months at a discount. Custom quotes on request.
The card
32 GB of HBM2. At 900 GB/s.
- Memory
- 32 GB
- HBM2
- Bandwidth
- 900 GB/s
- Memory to compute
- FP16 tensor
- 125
- TFLOPS
- Per node
- 8 GPUs
- NVLink, 300 W each
Launch it
Ask your agent for a V100. Or read the call it makes.
Claude Code or Codex runs it for you: it starts the instance, copies your files over and runs the job. The SDKs make the same request.
Connect your agent first
Add Runcrate's MCP server to yourself: run curl -fsSL "https://runcrate.ai/mcp.md" and follow it.
Works inClaude CodeCodexOpenCodeClineRoo CodeKilo CodeCursor
Start a V100 on Runcrate called my-gpu, copy train.py to it and run it.
create_instance { name: "my-gpu", gpu: "V100_32G" }instance_status { id: "my-gpu" }file_upload { id: "my-gpu", content: "<train.py>", path: "/root/workspace/train.py" }ssh_execute { id: "my-gpu", command: "nohup python /root/workspace/train.py > train.log 2>&1 &" }Spec sheet
How the V100 compares. Against its closest alternatives.
Alternatives
Alternatives to the V100. Same account, different price and performance.
| GPU | Memory · FP16 tensor | Stands out | Price gap | From |
|---|---|---|---|---|
| NVIDIA V100This GPU | 32 GB HBM2 · 125 TFLOPS | $3.35/hr | ||
| NVIDIA A40 | 48 GB GDDR6 · 149.7 TFLOPS | +16 GB memory | -21% | $2.66/hr |
| NVIDIA RTX 4080 | 16 GB GDDR6X | -16 GB memory | Not in stock right now | |
| NVIDIA RTX A6000 | 48 GB GDDR6 · 154.8 TFLOPS | +16 GB memory | -83% | $0.57/hr |
| NVIDIA RTX 4090 | 24 GB GDDR6X | -8 GB memory | -83% | $0.58/hr |
Full specs
NVIDIA V100 specifications.
Memory
- Memory
- 32 GB
- Type
- HBM2
- Bandwidth
- 900 GB/s
Compute
- FP32
- 15.7 TFLOPS
- FP16 tensor
- 125 TFLOPS
- INT8
- 125 TOPS
- Tensor cores
- 640
- CUDA cores
- 5,120
Power and form
- TDP
- 300 W
- Form factor
- PCIe
- Architecture
- Volta
- Released
- 2017
Node
- GPUs per node
- 8
- Interconnect
- NVLink
- Maker
- NVIDIA
Questions
Renting the V100. What buyers ask first.
Other GPUs
Browse every GPU by architecture.
NVIDIA Blackwell
NVIDIA Hopper
NVIDIA Ampere
NVIDIA Ada Lovelace