EC2, EBS, S3, VPC, IAM, on bare-metal, at the edge, or on a partner Neocloud. The Terraform, SDKs, and every AWS API call your software already makes keep working against a new endpoint.
381 AWS API actions, implemented against the real API rather than a mock. The AWS SDK, the AWS CLI and Terraform point at one endpoint, and your workloads run unchanged at the edge, on-premise, or on a partner Neocloud.
Roadmap items ship under the same AWS API surface. Code written for AWS today keeps working the moment they land. Track what's shipped in the release notes.
Each platform here has been through a Spinifex validation on the actual silicon, and the reference architecture and benchmark numbers are published in our docs. The list grows with every vendor we work through.
An AWS workload moving out to your own edge and on-premise sites lands here, on hardware your teams already rack and support. Three references are published end to end: the platform benchmark, a production vision pipeline, and vLLM serving measured on Intel AMX against NVIDIA L4.
Move GPU work off AWS without touching the stack that runs it, from training through inference and the general compute around them. Supermicro gives us the deepest accelerator coverage of any platform we support: NVIDIA H200 partitioned through MIG, RTX Pro 6000 serving under EKS, and AMD Instinct MI350X on the H14 chassis.
Carries the same AWS workload out to sites with no rack and nowhere to put one. A single fanless HX401 runs a complete Spinifex deployment on its own, with no machine room behind it.
Self-host the open-source build on your own hardware, or request a hosted environment on one of our Neocloud partners. One platform underneath, identical APIs either way.
Bare-metal, VM, or a multi-node cluster in your data centre. AGPL, yours to keep, no phone-home.
$ curl -fsSL https://install.mulgadc.com | bash
We provision a Spinifex environment on a partner Neocloud and hand you the credentials on the spot. Nothing to install, no card required, just an endpoint you can point your AWS CLI or Terraform at.
Credentials in under a minute.
Most AWS alternatives wrap an API and call it a cloud. We engineered the parts that are actually hard. That depth is the reason your software runs unchanged, instead of rewritten.
Spinifex serves the EC2, EBS, S3, VPC, and IAM APIs. Point your existing tooling at a new endpoint.
Every service runs on your hardware, with no dependency on a network or authority you don't control.
Spinifex manages VM lifecycles on QEMU with direct hardware access, exposed through an EC2-compatible API.
Predastore for erasure-coded object storage, Viperblock for replicated block volumes with NVMe caching.
Built for disconnected, contested, and resource-constrained environments. Connectivity is optional.
The source is public. A commercial license is available for deployment outside the terms of the AGPL.
The EC2, EBS, VPC and S3 APIs your software already speaks, behind a console your engineers can drive on day one.
Assess Spinifex against the AWS services and tools your workload uses today.
Read the guide →Plan a local deployment and understand what remains familiar and what you operate.
Read the guide →Evaluate Predastore, SigV4 clients, multipart uploads, and storage topology.
Read the guide →Spinifex is an open-source infrastructure platform that recreates the AWS service surface (EC2, EBS, S3, VPC, IAM) on bare-metal, edge, and on-premises environments. Run cloud-native software without a hyperscaler.
Yes. Drop-in compatible with the AWS CLI, the AWS SDKs, and Terraform. If your application talks to EC2, EBS, or S3 today, it talks to Spinifex tomorrow. No code changes required.
No. Designed to operate fully offline. Every service runs locally on your hardware with zero dependency on external networks or cloud control planes. Connectivity is an option you can take, never a dependency.
Any Linux-capable system, x86 or ARM. No proprietary hardware requirement. Designed for everything from rack-mounted servers to rugged edge nodes.
Two engines. Viperblock for high-performance block storage (EBS-compatible) with snapshots, replication, and NVMe caching. Predastore for S3-compatible object storage with Reed-Solomon erasure coding, multipart uploads, and multi-node clustering. If a node is lost, objects remain available from surviving nodes, with no manual intervention.
Yes. V1 is publicly available. Core compute, block storage, and object storage services are production-ready and deployed.
AGPL-3.0. The source is open and freely available. A commercial license is available for organisations that require deployment outside the terms of the AGPL.