bydtech-model-retire

Find deprecated, retiring and retired LLM model IDs in a codebase before a provider shuts them off.

This is the Python package. The same tool is published to npm as @bydtech/model-retire; both implementations ship the same dataset and produce identical output. The full documentation is in the repository README.

Install

pip install bydtech-model-retire

Python 3.9 or later. No runtime dependencies.

Quick start

model-retire .                      # scan the current directory
model-retire src --format json      # machine-readable output
model-retire list --provider openai # print the dataset for one provider

python -m model_retire works too.

Each hit is reported with file and line, the model, its status, the shutdown date, days remaining, the provider's recommended replacement and the official page the date came from. The exit code is 0 when nothing is at error level, 1 when something is, 2 for a usage error.

API

from model_retire import scan, format_text

result = scan(["src"], warn_days=60)
print(format_text(result), end="")
if result["summary"]["errors"]:
    raise SystemExit(1)

scan() takes paths, cwd, today, warn_days, warn_only, ignore, include, gitignore, include_lockfiles, dataset and max_bytes, and returns a dict with the same keys as the JSON output (version, today, warn_days, warn_only, files_scanned, summary, findings). Also exported: default_dataset, load_dataset_file, merge_datasets, validate_dataset, create_matcher, assess, format_text, format_json, format_github, format_list, walk_files and UsageError.

The data

Dates come from providers and can change; check the source link on each finding. Every entry in the dataset was read from the provider's own deprecation or lifecycle page and carries that page's URL. Corrections are welcome as issues or pull requests with the official source URL.

License

MIT. Maintained by BYD Tech, a software studio that builds MVPs, AI agents, and web & mobile apps.