Rent NVIDIA GTX 1080 Ti GPU by the hour
The NVIDIA GTX 1080 Ti is a consumer GPU built on the Pascal architecture with 11 GB of VRAM. It delivers about 11.0 TFLOPS of single-precision performance and about 339 GB/s of memory bandwidth within a 250 W power envelope. You can rent this GPU on ONIE with per-hour billing, SSH access, and ready-to-use software templates.
Specifications
VRAM per GPU
Architecture
Compute capability
FP32 compute
Memory bandwidth
Power limit
Values reported by the infrastructure provider for machines currently listed; real performance varies by machine.
Available configurations
Live stock and prices, updated continuously. The final price is confirmed before you order.
Loading live prices…
What the GTX 1080 Ti is good for
Prototyping and Code Testing
Test and debug PyTorch/CUDA scripts on a Pascal architecture card with compute capability 6.1.
Lightweight Model Inference
Run small language models through Ollama or vLLM templates that fit inside the 11 GB memory limit.
Standard Image Generation
Use the ComfyUI template to run basic image synthesis pipelines that do not require modern tensor cores.
Classic Deep Learning Training
Train foundational neural networks and computer vision models using about 11.0 TFLOPS of single-precision compute.
Is it the right GPU for you?
Choose the GTX 1080 Ti when you need 11 GB of VRAM for legacy workloads, code testing, or small-scale inference on a Pascal-generation consumer card. Select a newer GPU with more memory for large models or modern mixed-precision workloads, or a smaller consumer GPU if your task requires minimal VRAM.
Compare with similar GPUs
Frequently asked questions
Is the GTX 1080 Ti suitable for modern LLM inference and fine-tuning?
With 11 GB of VRAM and Pascal architecture at compute capability 6.1, it can run quantized small language models using templates like Ollama. For larger models or workloads that rely on newer architecture features, a more modern GPU with higher VRAM is recommended.
What software templates are available on ONIE for this GPU?
ONIE provides ready templates including PyTorch/CUDA, vLLM, Ollama, and ComfyUI. You also get full SSH access to configure your own environment.
How are performance and memory bandwidth measured for this card?
The infrastructure provider reports about 11.0 TFLOPS of single-precision compute and measures memory bandwidth at about 339 GB/s. The card operates at a 250 W thermal design power.
How does billing work when renting the GTX 1080 Ti?
Rentals on ONIE use per-hour billing, allowing you to pay only for the time your instance is active.
Rent the GTX 1080 Ti by the hour
Create an account, choose a configuration and connect over SSH. Billing stops when you delete the GPU.
Rent GTX 1080 Ti