A familiar interface
We intend to build on Guild’s existing local API capabilities, making integration with applications a central design goal.
Illatio Cloud / in development
Beyond your
own machine.
Local inference puts you in control. Hosted inference should make it easier to get started. Illatio Cloud is our planned platform for bringing Guild’s systems approach to a managed environment.
Platform Status In Development
Local Engine Available Today
Wire Format OpenAI / Anthropic
Isolation Session-Scoped
Heterogeneous MoE execution pipeline Conceptual diagram · measured metrics above
01 / Product direction Less infrastructure
Less infrastructure
between you and an idea.
Your Machine
Guild local runtime · available today →
Shared Contracts
GPACK containers & local API →
Illatio Cloud
Managed platform · in development
Hardware-aware execution
The starting point is Guild’s work on heterogeneous memory, portable model packages, and machine-specific optimization.
Visible boundaries
Model compatibility, resource limits, and measured performance should be clear before you depend on a service.
02 / Where things stand Building the foundation.
Building the foundation.
Not yet a service.
Illatio Cloud is in development. Public inference endpoints, API keys, pricing, launch dates, and service-level commitments are not available yet.
For inference you can run today, Guild is open source. Build the engine, choose a compatible model, and serve it on your own hardware.
Read the self-hosting guide- Guild inference engine Available source
- Local API serving Available in Guild
- Managed hosted inference In development