release: 1.20.0

[stainless-app]

Automated Release PR

1.20.0 (2026-01-17)

Full Changelog: v1.19.0...v1.20.0

Features

• client: add support for binary request streaming (b2af547)

Chores

• internal: update actions/checkout version (04aae72)

---

This pull request is managed by Stainless's GitHub App.

The semver version number is based on included commit messages. Alternatively, you can manually set the version number in the title of this pull request.

For a better experience, it is recommended to use either rebase-merge or squash-merge when merging this pull request.

🔗 Stainless website
📚 Read the docs
🙋 Reach out for help or questions

[cursor]

PR Summary

Introduces first-class binary request bodies and streaming.

• Add content support to post/patch/put/delete (sync/async); reject mixing content with json_data/files and warn when passing raw bytes via body (deprecated)
• Extend _build_request to send options.content; improve debug logging for Pydantic v1
• Add BinaryTypes/AsyncBinaryTypes; include content in FinalRequestOptions
• New tests for binary uploads (bytes, iterators, async iterators) and deprecation coverage
• Bump actions/checkout to v6 in CI workflows; update version to 1.20.0 and changelog

^{Written by Cursor Bugbot for commit e650197. This will update automatically on new commits. Configure here.}

[stainless-app]

🧪 Testing

To try out this version of the SDK, run:

pip install 'https://pkg.stainless.com/s/luma_ai-python/04aae7237e2b8289153bcf576ccc0882728efdf8/lumaai-1.19.0-py3-none-any.whl'

Expires at: Mon, 16 Feb 2026 05:44:58 GMT
Updated at: Sat, 17 Jan 2026 05:44:58 GMT

[cursor]

Iterator content exhausted on retry sends empty body

High Severity

When content is an Iterable[bytes] (such as a generator), the retry logic fails silently. The model_copy at lines 972 and 982 performs a shallow copy, so the same iterator reference is shared across all retry attempts. After the first request consumes the iterator, subsequent retries send an empty body because the iterator is exhausted. With DEFAULT_MAX_RETRIES=2, this affects users by default when requests fail with 429, 500, or timeout errors.

Additional Locations (1)

• src/lumaai/_types.py#L64-L66