Icosa Computing

Local AI that learns the work.

Everything you love about your AI assistant, without usage limits or bills that grow out of control. Icosa runs on your computer, not in the cloud.

  • No token costs. Use it as much as you want.
  • Super fast. Answers instantly, right on your device.
  • Private. Your files never leave your computer.

Trusted by

FujitsuNECToshibaNASANational Science FoundationUSRAAlgoquantUNLU & CoBener Law OfficeDualityCreative Destruction LabFujitsuNECToshibaNASANational Science FoundationUSRAAlgoquantUNLU & CoBener Law OfficeDualityCreative Destruction LabFujitsuNECToshibaNASANational Science FoundationUSRAAlgoquantUNLU & CoBener Law OfficeDualityCreative Destruction LabFujitsuNECToshibaNASANational Science FoundationUSRAAlgoquantUNLU & CoBener Law OfficeDualityCreative Destruction Lab
$0
Token costs
100%
Compute on your hardware
$1.4M
NSF research funding →

Products

Zeno

macOS

The local AI workspace that learns your work. Zeno adapts to your documents, code, and terminology, entirely on your machine, with no model selection or setup.

Apply for early access

Launching August 19

--days
--hrs
--min
--sec

LM Shop

Web

The no-code platform for training your own AI models. Upload your documents and get a custom model out, fast. Download it and run it locally with llama.cpp or any app you like, or directly with Zeno.

Use LM Shop

How it works

  1. 1. Upload your docs
  2. 2. Train with no code, in minutes
  3. 3. Download & run locally

Your hardware does the compute.
Usage never adds a bill.

01

Zero token costs

Agentic work turns one request into many model calls. On rented AI, that compounds a bill. On Icosa, it costs nothing.

02

Private by design

Files, prompts, and model weights never leave machines you control. No cloud dependency and no vendor shutdowns.

03

Expert accuracy

Combinatorial Reasoning scores many answer paths and selects the strongest, closing the gap with hosted frontier models.

Research

Many reasoning paths. One better answer.

Combinatorial Reasoning is a physics-based optimization method developed with NASA and funded by the National Science Foundation. It samples many reasoning traces, scores them, and selects the strongest set before answering, so small local models perform like much larger ones on the work they know.

Combinatorial Reasoning pipeline from the Icosa paper: sampled reasons are mapped to a QUBO problem, optimized, and the selected reasons feed the final prompt

Esencan et al., arXiv:2407.00071

Team

Deep-tech researchers, shipping product.

Mert Esencan

Founder & CEO

Stanford BS/MS, Oxford quantum PhD. Ex-Fidelity researcher.

Can Unlu

Software

UVA Math & CS. Founding engineer of the optimization stack.

Tarun Advaith

Research

Waterloo PhD. Drives Icosa research; author of ML and quantum papers.

Baran Melik

Go-to-Market

Duke MBA. Drives go-to-market and revenue operations.

Alan Ho

Product

Ex-Google Head of Quantum AI Product. Co-founder of Qolab.

Mark Gurvis

Revenue

Ex-Google Head of Strategic Platforms. Scales revenue.

AI shouldn't be rented by the token.

Bring owned, private AI to your team, on the hardware you already have.