Hey Hugging Face community 
I’m excited to share Daugherty Engine, a GPU-accelerated constraint satisfaction engine delivering quantum-competitive performance on standard classical hardware — and now validated at a 1,000,000-variable scale.
We recently completed a 1M-spin Ising optimization run, live and publicly verifiable, representing a problem size long considered the practical domain of quantum hardware. No clusters. No cryogenics. No specialized infrastructure.
Hugging Face Space:
https://huggingface.co/spaces/GotThatData/daugherty-engine
What it does
Daugherty Engine operates directly in the hardest regimes of combinatorial optimization.
It includes a 3-SAT solver tested at the phase transition (α ≈ 4.27), the empirically hardest region for boolean satisfiability.
It also includes an Ising model optimizer for spin-glass energy minimization — the native problem class targeted by quantum annealers like D-Wave.
Why this matters
The 1M-variable run isn’t just a scale demo — it’s an economics and physics result.
Here’s the contrast:
• Daugherty Engine: ~195W, ~$1.57/hour
• D-Wave Advantage: ~25kW, ~$13.20/hour
• IBM Quantum: ~15kW, ~$1.60/hour
That’s roughly 128× less power consumption than quantum annealing systems for the same class of problems, while remaining fully classical, transparent, and reproducible.
This is about changing the math of optimization, not escalating the hardware arms race.
Try it
The Hugging Face Space calls our public API — no proprietary source code is exposed.
You can:
Run SAT verification (20–500 variables)
Run Ising optimization (10–500 spins)
View real-time energy, cost, and performance comparisons
Tech stack
Backend: NVIDIA RTX 6000 Ada (48GB VRAM)
Frontend: Gradio
API: Flask on a GPU-enabled DigitalOcean instance
If you’re working in optimization, SAT, Ising models, or quantum-inspired computation, I’d love your feedback and critical review.
Contact: Shawn@smartledger.solutions
Live 1M-variable benchmark: https://1millionspins.originneural.aiNote: The Show and Tell category is for sharing and discussing projects, showcasing your Spaces, Models, Datasets and more. We value open-source and technical details over promotional content, so focus on sharing the intricate aspects of your work.