About Cogerphere AI Labs
We're Cogerphere
Intelligence in Motion. Knowledge in a Sphere. We exist to expand what systems can understand, retain, and uphold — so the future stays coherent.
Why We Exist
Cogerphere was founded on a simple belief: intelligence should move, and knowledge should gather. As AI systems grow in capability, they also grow in context — and that context is where coherence breaks. We build tools and research that keep the sphere intact.
From Cobentt — documents, slides, sheets, deep research, chat, websites, and smart contracts in one AI workspace — to Openbentt, our local-first research workspace, RRbench for research verification, and Meridian 0.1, our published LaTeX-aware academic proofreading SLM, we ship systems that think in context, scale with integrity, and stay human-aligned.
What we're building
Cobentt
One AI workspace for documents, slides, sheets, deep research, chat, websites, and smart contracts. Everything you make, in one place at cobentt.vercel.app.
cobentt.vercel.appOpenbentt
A local-first AI workspace for LaTeX, PDFs, benchmarking, and fine-tuned small models. Your research stays on your machine — no cloud accounts, no hidden logging.
openbentt.cogerphere.comRRbench
ResearchBench — a trust-first platform for assessing credibility and reproducibility of scientific research. Multi-signal Trust Scores combine AI claim analysis, cryptographic artifact checks, expert review, and community validation.
Read the previewMeridian 0.1
Our published LaTeX-aware academic proofreading SLM. The paper — Why Data Design Beats Method Innovation — introduces Meridian-AC-Nano, a 751.6M-parameter multi-task academic assistant. Runs on your hardware.
Read the full reportChillarOps
Enterprise-grade cybersecurity — offensive security, cloud security, DevSecOps, and secure infrastructure. Security by design with 24/7 support from certified engineers.
Visit ChillarOpsHow We Work
Research-first
We start with hard problems. Context drift, coherence degradation, security gaps — we tackle them before they scale.
Safety by default
Every system we build is designed to fail safely. No shortcuts, no silent drift.
Transparent by design
We share what we learn. Open-source where it matters, clear documentation always.
Want to expand the sphere with us?
About · Cogerphere AI Labs — research, products, and human-aligned intelligent systems.
← Home