Kentucky AI Shelbyville · Kentucky

Kentucky AI.

Research & Development

An independent AI research-and-development lab in Shelbyville, Kentucky.
We fine-tune, quantize, and serve domain-adapted model stacks — and the inference infrastructure beneath them — for operators across Kentucky.

At capacity · waitlist open · submissions reviewed personally

View the registry Join the waitlist
The registry

Model cardsWhat's on the bench.

Every model ships with a card — base weights, quantization, license, pipeline, status. Some in production, some in training, some commissioned, some on the bench.

In production
Spline · 2026

Document-grounded quantity takeoff.

A fine-tuned, tool-using model for commercial preconstruction. Parses plansets and specifications, runs quantity takeoff and scope extraction, prices against a historical cost index, and flags substrate and moisture risk — every figure traced to a source line by span-level attribution.

License
Apache 2.0
Base model
Qwen2.5-14B-Instruct
Weights
14B · 4-bit GPTQ
Tensors
Safetensors · BF16
Pipeline
Text-to-Text · tool-calling
Status
In production
View on GitHub
In research
Bloodstock · 2026

Multimodal phenotype regression.

A multimodal regressor over Thoroughbred performance, pedigree, conformation, and biometric tensors — vision and tabular encoders fused into a shared latent. Emits calibrated win/place posteriors, not narratives.

License
Apache 2.0
Base model
Qwen2.5-VL-7B-Instruct
Weights
7B · BF16
Tensors
Safetensors · FP16
Pipeline
Image-Text-to-Text · tabular fusion
Status
In research
View on GitHub
Commissioned
Environmental · 2026

Classifier-routed exposure triage.

An agentic abatement workflow commissioned by AIM Air Monitoring & Asbestos Testing. Each intake routes through a tuned sequence classifier and a reasoning agent that scores asbestos-exposure and regulatory risk against operator-set thresholds, with deterministic escalation paths.

License
MIT
Base model
Phi-4
Weights
14B · 4-bit
Tensors
Safetensors · BF16
Pipeline
Seq-classification · agentic routing
Status
In build
View on GitHub
In testing
NVIDIA · LocateAnything-3B

Parallel-decode visual grounding.

A vision-language grounding model from NVIDIA, on our bench. Open-set localization, dense detection, and point-level grounding via Parallel Box Decoding — full bounding-box coordinates in one forward pass. Under evaluation for plan- and document-layout grounding.

License
NVIDIA · non-commercial
Base model
Qwen2.5-3B-Instruct
Weights
3B · BF16
Tensors
Safetensors · BF16
Pipeline
Image-Text-to-Text
Status
In testing
View on GitHub Model on Hugging Face ↗

More cards as they ship.

Live bench

Try itThe model reads the plan.

Not screenshots. A real architectural floor plan read by NVIDIA's LocateAnything-3B on our bench — rooms, areas, finish tags, and notes located and transcribed in a single forward pass. Hover any box to see exactly what the model saw. Plus live vision Spaces you can drive with your own image, right here in the page.

LocateAnything-3B · reading the sheet
Architectural planset with model-detected elements
Hover a detection
Model nvidia/LocateAnything-3B · parallel box decoding · run locally on Apple Silicon Sheet: architectural finish plan A7-1 (used with permission) · evaluation only · non-commercial
NVIDIA · LocateAnything — live Open in new tab ↗
Automated floor-plan digitization — live Open in new tab ↗

Live Spaces run on shared GPUs — a cold model may queue for a moment.

Capabilities

The practiceIf it's worth building,
we build it.

An independent R&D lab for applied AI. We architect, train, and operate models — and the unglamorous infrastructure that turns a clever demo into something that survives production.

Foundry

Model training & fine-tuning.

We forge open-weight foundations — Llama, Qwen, Mistral — into domain-native models: continued pretraining, LoRA and full-parameter fine-tunes, distillation, quantization, shipped to inference.

Inference

Private, on-prem deployment.

Tuned open models served behind your firewall — air-gapped and sovereign. Weights, latency, and corpus stay in the building, under your control.

Compute

Training compute, by the run.

Reserve local GPU capacity for fine-tunes and large jobs — orchestration, checkpointing, and the eval harness handled, with an architect on the line.

Corpora

Data acquisition & curation.

Web-scale acquisition, extraction, deduplication, and structuring — turning the open web and your archives into clean, model-ready corpora.

Architecture

AI-readiness assessments.

We map your stack and return a blueprint: where agents earn their keep, where they don't, and the infrastructure to run them safely.

Research

Commissioned research & systems.

Bespoke agentic pipelines, retrieval, and evaluation — original work built for a single question or a single operator.

At capacity · waitlist open · email me, I review every submission →

About

We ship weights,
not decks.

Kentucky AI is an independent AI research-and-development lab in Shelbyville, Kentucky. We fine-tune, distill, quantize, and serve domain-adapted model stacks — and the inference infrastructure beneath them — for a small roster of operators and for our own research. The thesis is plain: capable, sovereign, locally-served models in the hands of people doing real work.

We architect, train, quantize, deploy, and operate — then hand over the methodology and the checkpoint, not a login. No SaaS seat. No consultancy deck. A lab that ships weights.

You're welcome to email. We're at capacity with a waitlist open — I review every submission personally.

Email me →