A small code-action agent that forecasts a project's final cost (EAC) and finish period from a mid-flight Earned Value Management snapshot. It is openbmb/MiniCPM5-1B (1B, Llama-style dense decoder) fine-tuned (LoRA, then merged) to run a single-tool reasoning loop: it writes Python that calls a curated forecasting toolkit (Earned Schedule, CPI/SPI formulas, a Gompertz growth curve, a reference-class ML regressor, and the TimesFM / Chronos time-series foundation models), reconciles their disagreeing estimates, and submits one answer.

It was distilled from a DeepSeek V4 teacher: the teacher's reasoning traces over a diverse simulated project corpus were filtered to a 367-trace SFT set (build-small-hackathon/slipstream-evm-sft) and the student trained with assistant-only loss (reasoning + tool-call tokens only). This makes a sub-5B, edge / air-gapped forecaster that matches the classical project-controls baseline and approaches its cloud teacher.

Scored on 107 real completed projects (Batselier/OR-AS DSLIB), apples-to-apples with every baseline. valid = produced a usable forecast; EAC error = median absolute % error on final cost; finish error = median absolute error in periods.

Distillation lifts a base model that could barely operate the tool-call format into a reliable forecaster that rivals the classical canon and its own teacher.

The model is trained to act through a single run_python(code=...) tool call and to call submit(finish, eac) from inside that code. See the Slipstream project for the agent loop, the forecasting toolkit, and the full benchmark.

This is a derivative of openbmb/MiniCPM5-1B and is released under the base model's licence (minicpm-model-license). You must comply with the upstream terms (licence). Training data: build-small-hackathon/slipstream-evm-sft.

Built for the Hugging Face Build Small hackathon.