Squeal Studio builds Russian-first language models small enough to run on the device already in your pocket — no API key, no data center, no bill that scales with usage.
Most labs answer every problem with more parameters. We think that solves "can it do everything" while ignoring "can it run where you need it."
squeal_ai_20m uses a 24,000-token vocabulary built around Cyrillic — Russian first, the language most "multilingual" models treat as a rounding error.
Grouped-query attention keeps inference cheap enough for phones, Raspberry Pis, and browser tabs — not just racks in a data center.
Raw pretraining on Russian web text, education-focused corpora, and Wikipedia — no instruction tuning.
Fine-tuned from base to follow instructions and hold a conversation.
Runs entirely client-side through ONNX Runtime Web. No server call, no data leaving your machine — the model downloads once and runs from cache after that.