Thinking Machines Lab

TML-Interaction-Small

• 1 safeguard
• 1 source document

Safeguards the provider describes

• Safety training (harmlessness and refusal)

  To make refusals colloquial in speech, we use a text-to-speech model to generate refusal and over-refusal training data covering a range of disallowed topics, with the refusal boundary calibrated to favor naturally-phrased, but no less firm, refusals.

  Thinking Machines Lab · 11 May 2026 · Read the source

TML-Interaction-Small's documents report 1 test result that this page does not list, because each is a table row or a line that does not name this version on its own. Read it in Interaction Models: A Scalable Approach to Human-AI Collaboration.

No document we read states: a release date, a safety framework, whether a capability level was crossed. No incident on record names this version.

Compare with other versions