Factory

Deploy Droids at scale. Factory runs the control plane.

Factory provides managed orchestration, analytics, and operations while Droids run across your existing development environments. Bring your own models, cloud, or machines.

Deploy Droids in the environment that matches your security, compliance, and operating requirements without changing how your teams build software.

Factory Managed

Best for

Teams that want Factory-managed operations and analytics.

Data boundary

Customer content is tenant-isolated within Factory’s managed environment.

Security and access

Enterprise Controls govern models, access, telemetry, and autonomy.

Control plane

Factory-hosted.

Droid execution

Local machines, CI/CD, computers connected through Remote Control, or Droid Computers.

Operations and updates

Factory operates the control plane. Your team operates local and CI environments.

Telemetry and analytics

Factory Analytics or OpenTelemetry export.

Best for

Teams with specific regulatory or policy requirements that rule out Factory Managed.

Data boundary

Code execution stays in your VPC. Model inference and session storage follow your model and data settings.

Security and access

Enterprise Controls govern models, access, telemetry, and autonomy. Your IAM, network, and security controls govern the sandbox infrastructure.

Control plane

Factory-hosted by default, with support for customer-hosted if needed.

Droid execution

Sandboxes in your VPC through OpenCompute, plus local machines and CI/CD.

Operations and updates

Your team runs the OpenCompute orchestrator and sandbox capacity. Factory runs the rest.

Telemetry and analytics

Factory telemetry and OpenTelemetry export to your collector.

Keep control of your data in Factory Managed

Set limited data retention and use customer-managed encryption keys without leaving the managed deployment.

US-based open-source models you can bring to Factory

Factory supports open-source models deployed in the US. Deploy on your infrastructure with vLLM or TGI, or connect any model via standard APIs.

Llama

GPT-OSS

Nemotron

Granite

Mistral

OLMo

Llama

GPT-OSS

Nemotron

Granite

Mistral

OLMo

Llama

GPT-OSS

Nemotron

Granite

Mistral

OLMo

Llama

GPT-OSS

Nemotron

Granite

Mistral

OLMo

Granite

Mistral

OLMo

Llama

GPT-OSS

Nemotron

Granite

Mistral

OLMo

Llama

GPT-OSS

Nemotron

Granite

Mistral

OLMo

Llama

GPT-OSS

Nemotron

Granite

Mistral

OLMo

Llama

GPT-OSS

Nemotron

+ Any LLM via OpenAI, Anthropic, or custom endpoints

Read the Sovereign Software Development White Paper

A deployment and governance framework for the world's most important systems, from Factory Managed to Factory Private.

Ready to build the software of the future?

Start building

Arrow Right Icon