Community Mac
Apple Silicon with host-native MLX for local private AI, MLX embeddings, and Srasta core planes.
Free Community Edition
Community Edition gives builders a real Srasta deployment: private inference, admin basics, governance basics, memory, governed tools, and audit evidence. Apple Silicon MLX is the first complete Community proof path; single NVIDIA GPU with vLLM is the next beta path.
Community path
Apple Silicon with host-native MLX for local private AI, MLX embeddings, and Srasta core planes.
Single compatible NVIDIA GPU with vLLM-backed private inference. Beta path after the Apple proof, no production SLA.
CPU-only runs may support internal wiring checks, but they are not a current public private-inference Community profile.
What is included
It is the real Srasta core platform on one node, with enterprise deployment features gated until a customer needs paid support, certified topology, or production guarantees.
What Srasta learns
Community deployments help Srasta certify real hardware, improve install quality, prioritize roadmap demand, and identify where users need Enterprise.
Install stage, OS, arch, RAM, disk, accelerator family, selected runtime, model/catalog ID, health status, error category, feature counts, and upgrade-interest events.
Prompts, responses, documents, file names, embeddings, retrieved context, secrets, audit-log payloads, end-user identities, and customer business data.
When to upgrade
Paid pilot or departmental deployment when one workflow needs support, evidence, and conversion planning.
Production private AI with certified deployment, broader user/workflow scope, governed memory/tools, support, and audit package.
Multiple environments, custom certified images, private registry, premium SLA, advanced compliance, and priority roadmap.