Tuned Tensor

Behaviour control infrastructure for open-weight language models.

Specify behaviour, fine-tune open-weight models, and use evaluation feedback to start the next run.

Specify behaviour: capture system prompts, guidelines, constraints, examples, and a base model in a structured spec.
Fine-tune: compile training data and tune supported open-weight models.
Evaluate: score outputs, inspect failures, and catch regressions.
Iterate: use run feedback to refine the spec and start the next tuning pass.

npm i -g @tuned-tensor/cli
tt init
tt runs start <spec-id>