Language

Domain LLMs, and the products that run on them

We adapt reasoning and coding models to a specialist corpus, compress them for a single datacenter GPU, and ship chat and analysis products that never need a third-party API.

See Thinking-LQ-1.0 Request a custom build

Thinking-LQ-1.0 — live inference MedQA · holdout

62-year-old, crushing chest pain radiating to the left arm. ST-elevation in leads II, III, aVF. Which artery is most likely occluded?

84%

MedQA accuracy

~20GB

checkpoint size

1.6×

faster than the 32B base

open checkpoints — Thinking-LQ & Coder-LQ

third-party APIs — private, on one L40

Models

Chaperone-Thinking-LQ-1.0

Open reasoning model: GPTQ + QLoRA on DeepSeek-R1-Distill-Qwen-32B. Medical and scientific corpora. Fully available on Hugging Face.

View results →

Chaperone-Coder-LQ-1.0

Quantized coding assistant for debugging and production snippets. Same deployability story: small enough for a single L40/L40s.

Hugging Face →

Applications

How the language line shows up for teams who never want to touch a checkpoint.

Chatbots

Conversational assistants on your data, privately hosted. Support teams stay on hard tickets; the model handles the rest.

Explore chatbots →

Specificity

Semantic analysis on specialist text — trends, sentiment, and extraction powered by Thinking-LQ, not a generic API.

Explore specificity →

CHAPERONE AI