NVIDIA · Ampere · 2020
NVIDIA A40. Rental by the hour, on demand.
From $2.66/hr per GPU on demand, or prepay one GPU at $2.53/GPU-hr for 1 month, or a 4-GPU machine at $2.45/GPU-hr for 3 months.
A40 on demand
- Availability
- Available now · 2 regions
- Prepaid
- 1 mo $2.53/hr3 mo $2.45/hr6 mo $2.37/hr12 mo $2.27/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
48 GB of GDDR6. At 696 GB/s.
- Memory
- 48 GB
- GDDR6
- Bandwidth
- 696 GB/s
- Memory to compute
- FP16 tensor
- 149.7
- TFLOPS
- Per node
- 4 GPUs
- PCIe, 300 W each
Price across clouds
NVIDIA A40, cloud by cloud. Sorted by price.
On-demand rates per GPU-hour from each cloud's public price page on 8 Nov 2026: RunPod's Secure Cloud, and Lambda's lowest per-GPU rate across its cluster sizes. Runcrate's row is today's live starting rate.
Launch it
Ask your agent for an A40. 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.
Start an A40 on Runcrate called my-gpu, copy train.py to it and run it.
create_instance { name: "my-gpu", gpu: "A40" }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 A40 compares. Against its closest alternatives.
Alternatives
Alternatives to the A40. Same account, different price and performance.
| GPU | Memory · FP16 tensor | Stands out | Price gap | From |
|---|---|---|---|---|
| NVIDIA A40This GPU | 48 GB GDDR6 · 149.7 TFLOPS | $2.66/hr | ||
| NVIDIA RTX A6000 | 48 GB GDDR6 · 154.8 TFLOPS | -79% | $0.57/hr | |
| NVIDIA RTX 4090 | 24 GB GDDR6X | -24 GB memory | -78% | $0.58/hr |
| NVIDIA V100 | 32 GB HBM2 · 125 TFLOPS | -16 GB memory | +26% | $3.35/hr |
| AMD MI210 | 64 GB HBM2e · 181 TFLOPS | +16 GB memory | Not in stock right now |
Full specs
NVIDIA A40 specifications.
Memory
- Memory
- 48 GB
- Type
- GDDR6
- Bandwidth
- 696 GB/s
Compute
- FP32
- 37.4 TFLOPS
- FP16 tensor
- 149.7 TFLOPS
- INT8
- 299 TOPS
- Tensor cores
- 336
- CUDA cores
- 10,752
Power and form
- TDP
- 300 W
- Form factor
- PCIe
- Architecture
- Ampere
- Released
- 2020
Node
- GPUs per node
- 4
- Interconnect
- PCIe
- Maker
- NVIDIA
Questions
Renting the A40. What buyers ask first.
Other GPUs
Browse every GPU by architecture.
NVIDIA Blackwell
NVIDIA Hopper
NVIDIA Ampere
NVIDIA Ada Lovelace
NVIDIA Volta