Metadata-Version: 2.4
Name: adapterci
Version: 0.1.0
Summary: Regression CI for fine-tuned local models. Scores travel, data doesn't.
License: MIT
Requires-Python: >=3.10
Requires-Dist: httpx>=0.27
Requires-Dist: pyyaml>=6.0
Description-Content-Type: text/markdown

# AdapterCI

Regression CI for fine-tuned local models. Runs entirely against your own
OpenAI-compatible endpoint (Ollama, LM Studio, llama.cpp, vLLM). Only metric
names and floats are ever sent to the AdapterCI API — no prompts, no
completions, no weights.

## Install

```bash
pip install adapterci
```

## Quickstart

```bash
adapterci init      # writes adapterci.yaml
# edit adapterci.yaml: api_key, project, endpoint, baseline_model, candidate_model
adapterci            # runs the suite, pushes scores, exits 1 on regression
```

## Golden sets

Add domain-specific test cases in a local JSONL file (never uploaded):

```json
{"prompt": "Summarize in one sentence: ...", "check": "contains", "expected": "..."}
```

`check` can be `contains`, `regex`, `numeric_answer`, or `json_schema_keys`.

## Pointing at a different API (staging/local testing)

```bash
export ADAPTERCI_API_BASE=http://localhost:8787
```

## License

MIT
