NVIDIA · Ampere · 2020
NVIDIA RTX 3090. Rental by the hour, on demand.
From $0.41/hr per GPU, pay as you go with no commitment. 24 GB of GDDR6X and 71 TFLOPS FP16 non-tensor, for training and inference.
RTX 3090 on demand
- Availability
- Available now · 1 region
- By region and size
- $0.41 to $0.42/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
24 GB of GDDR6X. At 936 GB/s.
- Memory
- 24 GB
- GDDR6X
- Bandwidth
- 936 GB/s
- Memory to compute
- FP16
- 71
- non-tensor TFLOPS
- Per node
- 2 GPUs
- NVLink, 350 W each
Price across clouds
NVIDIA RTX 3090, 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 RTX 3090. 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 RTX 3090 on Runcrate called my-gpu, copy train.py to it and run it.
create_instance { name: "my-gpu", gpu: "RTX3090" }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 RTX 3090 compares. Against its closest alternatives.
Alternatives
Alternatives to the RTX 3090. Same account, different price and performance.
| GPU | Memory · FP16 non-tensor | Stands out | Price gap | From |
|---|---|---|---|---|
| NVIDIA RTX 3090This GPU | 24 GB GDDR6X · 71 TFLOPS | $0.41/hr | ||
| NVIDIA RTX 4070 Ti | 12 GB GDDR6X · 80.2 TFLOPS | -12 GB memory | Not in stock right now | |
| NVIDIA RTX 4080 | 16 GB GDDR6X · 97.5 TFLOPS | -8 GB memory | Not in stock right now | |
| NVIDIA V100 | 32 GB HBM2 | +8 GB memory | +717% | $3.35/hr |
| NVIDIA A40 | 48 GB GDDR6 | +24 GB memory | +549% | $2.66/hr |
Full specs
NVIDIA RTX 3090 specifications.
Memory
- Memory
- 24 GB
- Type
- GDDR6X
- Bandwidth
- 936 GB/s
Compute
- FP32
- 35.6 TFLOPS
- FP16 non-tensor
- 71 TFLOPS
- INT8
- 284 TOPS
- CUDA cores
- 10,496
Power and form
- TDP
- 350 W
- Form factor
- Consumer
- Architecture
- Ampere
- Released
- 2020
Node
- GPUs per node
- 2
- Interconnect
- NVLink
- Maker
- NVIDIA
Questions
Renting the RTX 3090. What buyers ask first.
Other GPUs
Browse every GPU by architecture.
NVIDIA Blackwell
NVIDIA Hopper
NVIDIA Ampere
NVIDIA Ada Lovelace
NVIDIA Volta