Fly.io

Fly.io offered a range of GPU machines suited for various compute-intensive applications such as inference, model training, and high-precision computations. However, these services are being discontinued and will be unavailable after August 1.

Fly.io

Provider Profile

Founded

Unknown

Headquarters

Unknown

Pricing Model

Specific GPU configurations had unique pricing, likely dependent on the model and region.

Technical Specification

Target Audience
Developers and businesses needing GPU power for applications such as ML model training and scientific computations
GPU Clusters & Offerings
  • NVIDIA A10
  • L40S
  • NVIDIA A100 40G PCIe
  • NVIDIA A100 80G SXM
Network Fabric
Cloud Hypervisor instead of Firecracker for GPU-enabled machines
Connectivity Bandwidth
Unknown
Storage Architecture
Ephemeral storage with a capacity of up to 50GB, Fly Volumes up to 500GB
Compute Framework Compatibility
CUDA based applications which likely include popular ML frameworks like PyTorch and TensorFlow
Resource Orchestration
Docker based deployment
Resource Availability
  • Deprecated
  • services ending after August 1
Datacenter Locations
Key Platform Features
  • Single full GPU per machine
  • Support for CUDA and ML applications
  • Tensor cores, RT cores, NVENC/NVDEC support

Last Audit: February 2026