Introducing BloxWeaver Sùil - fast, multilingual, multimodal decisions

Some decisions depend on what you can see, not just what you can read. That’s why we built BloxWeaver Sùil.

Sùil (“sool”, Scottish Gaelic for “eye” or “glance”) is our own fast, multilingual System One model, calibrated for both vision and text-based decisions across languages.

Like JEV, send it a state and typed questions, and get back structured answers with probabilities. No text generation, no parsing. However, Sùil goes further by handling images natively alongside text, so you can ask questions about what’s in a document, product photo, or scanned form in the same call.

We’ve also updated Pipes to match. You can use the existing JEV connector, as the API is compatible with TypeSafe AI’s System One API, or use it from the BloxWeaver connector, which now has decision action support.

This means you can use Sùil for image-and-text decisions, or text-only, right alongside your existing JEV questions. Same primitives, same structured output, now with vision built in.

Fast, calibrated, multilingual, and multimodal. Available now via API or in Pipes.

Ready to see Sùil in action? Book a demo today and see how BloxWeaver can transform your processes.