Small Language Models

Compact, efficient language models (typically under 1B parameters) designed for accessibility, affordability, on-device inference, and educational purposes, as alternatives to large-scale frontier models.

Reading List

Under the Hood

The System is the Moat: Why Small Models Rival Frontier AI in Cybersecurity

Apr 11, 20261268

AI cybersecurity is a 'jagged frontier' where small models often match frontier performance, proving that the orchestration system is the true competitive moat.

Cybersecurity Small Language Models AI Benchmarks Competitive Moats Vulnerability Research

Products & Announcements

nanochat: Train and Serve a $100 Mini ChatGPT in 4 Hours

Oct 14, 20251523

Train and serve your own mini ChatGPT in ~4 hours for ~$100 with a single, minimal, end-to-end codebase—and scale it with a few simple tweaks.

LLM Training Open Source GPU Computing AI in Education Small Language Models

Agentic Systems

Prompted to Perform: A 22% Lift for GPT-5-mini on Tau² Telecom

Sep 17, 2025197

A structured prompt rewrite turned vague policies into checklists, boosting GPT-5-mini’s telecom benchmark accuracy by 22% and unlocking previously unsolvable tasks.

Prompt Engineering AI Benchmarks Small Language Models AI Agents