OV-engine

OV-Engine is a *Universal Hallucination-Free Inference Runtime*. C++ kernel to intervene in the model's internal layers (Attention Matrices & SSM States)

Comment

Inspiration

Om Vinayaka because of my health some days i cant talk sometimes I can’t type because I get puss boils all the times I have had androgen insensitivity syndrome so my body didn’t accept testosterone I became a woman even now when I sometimes get puss boils so i have to use ai all the time sometimes for speaking to text sometimes for text to speech and to write my research papers so when i used normal ai with rag it even then hallucinating every one trying to use rag to reduce hallucinations but the problem is model’s internal drift causes the hallucinations I want to catch that to prevent hallucinations so for helping me and all people I have built this

What it does

🌸 The Philosophy: Biology over Brute Force

For years, AI memory (RAG) has been treated like a flat list—a library where you must run down every aisle to find a book. It works, but it is heavy, inefficient, and prone to "hallucinations" when the data gets noisy.

OpenVinayaka takes a different approach. Inspired by nature, it gives AI a Metabolism.

Rest (Low Energy): It focuses only on what is vital (High Centrality). Explore (High Energy): It allows creativity but anchors it in Truth. By mathematically intervening in the model's internal state using the Priority Formula, we transform "Probability" into "Reliability".

P(d) = S (q,d) × C (d) × R (d) × W (d)

How we built it

I am a scientist I am not developer so I vibe coded using Gemini cli

Challenges we ran into

None because I do research first then I take action and but code Gemini cli took care of that

Accomplishments that we're proud of

We tested OV-Engine against Standard Vector RAG on 10,000 "Trap" scenarios designed to trick AI (e.g., Version Conflicts, Security Negation).

Metric Standard RAG OV-Engine Wins 1,063 10,000 Failures 8,937 0 Accuracy 10.6% 100.0% Throughput ~67 q/s ~67 q/s While standard RAG chases keywords (distractors), OV-Engine respects Structural Authority.

What we learned

We don't need more compute. We need better geometry.

What's next for OV-engine

I want to help all ai model providers reduce hallucinations because with hallucinations there are many people and organisations and all governments using ai with one mistake it could ruin many people

Built With

Share this project:

Updates