Tuned Tensor

Tuned Tensor Documentation

Tuned Tensor turns a behaviour spec into a fine-tuned small open-weight model and a paired report showing whether it beat the run baseline on representative evaluation cases.

The tt CLI is the primary interface. Run tt to open the interactive terminal, or call commands directly: tt local ... trains on your own NVIDIA GPU, plain tt (or tt cloud ...) uses managed Tuned Tensor. Both paths use tunedtensor.json as the behaviour spec.

Key Concepts

Behaviour Specs

A behaviour spec is a structured description of what you want your model to do. In CLI and local workflows, this lives in your project as tunedtensor.json. It includes:

  • System prompt — the persona and role of the model
  • Guidelines — rules the model should follow
  • Constraints — things the model must not do
  • Examples — input/output pairs demonstrating desired behaviour
  • Base model — which open-weight model to fine-tune

Runs

A run snapshots the behaviour spec, prepares training and evaluation data, fine-tunes the model, and compares the run baseline with the tuned candidate on representative cases. Local and managed runs differ in who operates the training infrastructure.

Paired Evaluation

Tuned Tensor evaluates the baseline and tuned candidate on matched representative cases and reports score and pass-rate deltas plus per-example regressions. The local workflow creates a holdout when the spec has enough examples; prebuilt data prefers validation and then test by default. Results are evidence on those cases, not a guarantee of production improvement.

Local Serving

Fine-tuned artifacts can be served locally with the tt CLI. Use tt models serve for a managed model or tt local models serve for a model trained through the local workflow.

Workflow

  1. Define — Create a behaviour spec describing what your model should do
  2. Run — Choose local or managed compute, evaluate the baseline, fine-tune the candidate, and evaluate it on matched cases
  3. Inspect — Review the paired report, aggregate deltas, and per-example regressions
  4. Serve — Use tt local models serve for a locally trained model or tt models serve for a managed artifact
  5. Iterate — Refine your spec and run again

Managed Base Models

Managed fine-tuning currently supports a restricted set of production base models. Prices below are per 1M training tokens, per epoch. See the Billing & credits page for details on how managed runs are charged.

ModelSizePrice / 1M tokens · epoch
google/gemma-4-E2B-itE2B$0.45
Qwen/Qwen3.5-2B2B$0.45
ibm-granite/granite-3.3-2b-instruct2B$0.45
Qwen/Qwen3-VL-2B-Instruct2B VL$0.55
meta-llama/Llama-3.2-3B-Instruct3B$0.55
bigcode/starcoder2-3b3B$0.55
google/gemma-4-E4B-itE4B$0.70
Qwen/Qwen3.5-4B4B$0.70
microsoft/Phi-4-mini-instruct3.8B$0.70

CLI & API

Both workflows are accessible through the open-source tt CLI. Managed workflows are also accessible through the REST API.

  • TT CLI — Install with npm install -g @tuned-tensor/cli. Open source (Apache-2.0) on GitHub. See the TT CLI docs for installation and usage.
  • REST API — Base URL: https://tunedtensor.com/api/v1. All endpoints accept Authorization: Bearer <api-key> API key authentication or session cookies from the dashboard.