All GPUs
NVIDIA · Ampere
NVIDIA A100 80GB
A cost-efficient, widely supported choice for training, fine-tuning, and inference.
- On-demand
- 2 years
- 3 years
Specifications
- Architecture
- Ampere
- GPU memory
- 80 GB HBM2e
- Memory bandwidth
- Up to 2 TB/s
- GPU-to-GPU
- NVLink 3 · 600 GB/s (SXM)
- Form factor
- SXM or PCIe
Best for
- Cost-efficient training and fine-tuning
- Inference for mid-size models
- Batch and research workloads