One-schema compile
Point the CLI at a Pydantic class or JSON Schema file; emit a .gbnf aimed at llama.cpp/Ollama.
Under Validation
A planned local CLI that compiles one schema file to llama.cpp GBNF and dry-runs it against a tiny local GGUF before you wire it into an agent loop. Waitlist only; no production binary yet.
Concept validation only. No payment, no commitment, no recurring newsletter.
Proposed workflow
Honest concept stage: none of the steps below ship as a product today. Signup measures interest only.
Point the CLI at a Pydantic class or JSON Schema file; emit a .gbnf aimed at llama.cpp/Ollama.
Run the grammar against a small local GGUF and surface the first failing token path before agent integration.
Schemas stay on your machine. No cloud upload of API shapes or proprietary models.
The honest status
The concept is being validated before development time is committed. Joining tells us the problem is relevant to you and gives you first access if the evidence supports a build.
Before you decide
The current scope, privacy model, and next step without launch-day promises.
No. This is a concept-validation page. A beta would be local-only if demand clears the bar.
Scope is intentionally narrower: one model/schema and a dry-run check, not full multi-path OpenAPI flattening.
No. The concept should state its limits clearly and preserve human review wherever judgment is required.
Your email is recorded for this experiment only. You receive one relevant update or beta invitation if the concept moves forward.
Early access
Join the waitlist. One email if a local beta opens; no newsletter drip.
Leave your email to receive the beta invitation if this concept moves forward.