Hey HN, Before we roll out our fully deterministic compiled language model (CLM), figured I'd post CIYA's v0.02 for you to try out. At this point CIYA acts as a storage and logic engine for any of your prompting needs, and is available via API/Websocket integration. If your company/project frequently repeats the same prompts, needs to template out those (or any) prompts, or wants to keep your business logic directly on site instead of giving it away to cloud providers, CIYA can do all of that for you.

Feel free to pull up the guide using the command R: guide to see the current list of available tools like the ability to run applications, model your own datasets and build your own independent response tables.

We've developed a clear path for CLM integration, allowing for a fully compiled language model that can resolve up to 1 million tokens (on-prem) in under 200ms without ever having to access an LLM.

Feel free to checkout our landing page to get in touch: https://iiio.co, or run R: contact in app to fill out the contact form.

Open to any feedback, comments or questions you may have!