Add an optional local embedding server

[IlyaKhalafi]

This PR adds fastembed serve for local text embeddings through an OpenAI-compatible endpoint. With this, clients can use a single loaded model over HTTP or load it from a local directory. Server dependencies stay optional.

Fixes #571

Tested with a real model and the OpenAI Python client, both CLI entry points and the built wheel. The 39 feature tests pass on Python 3.10, 3.12 and 3.14. The full suite passes with 236 tests passed and 13 skipped. Type checks and pre-commit checks pass.

All Submissions

• Followed the contribution guidelines
• Checked existing PRs for the same change

New Feature Submissions

• Existing test suite passes
• Added feature tests
• Pre-commit hooks are installed and pass

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

• Configuration used: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: 7ed2519f-46fb-492e-bc2b-94c7640ccb1b

📥 Commits

Reviewing files that changed from the base of the PR and between de502ca and ba47103.

⛔ Files ignored due to path filters: 1

• uv.lock is excluded by !**/*.lock

📒 Files selected for processing: 11

• .github/workflows/python-tests.yml
• .github/workflows/type-checkers.yml
• .github/workflows/weekly-tests.yml
• README.md
• docs/embedding_server.md
• fastembed/__main__.py
• fastembed/cli.py
• fastembed/server.py
• pyproject.toml
• tests/test_cli.py
• tests/test_server.py

Included review availability: This review used your included allowance. Your plan provides up to 8 included reviews per hour; 7 remain after this review.

📝 Walkthrough

The change adds an optional local HTTP embedding service with health, model-list, and embeddings endpoints. The fastembed serve command configures and starts the service with Uvicorn. The API supports float or base64 embeddings and validates requests and model output. Tests cover API behavior and CLI options. Documentation describes installation, configuration, endpoints, and deployment notes.

Priority: ➖ Normal

Estimated code review effort: 3 (Moderate) | ~20 minutes

Change: Feature

Merge Risk: ⚪ Minimal · up to ba471

The local embedding service is mergeable after normal checks; no outstanding issue was established.

Pre-merge checks: 4 · 1

Failed checks: 1 warning

Check name	Status	Explanation	Resolution
Docstring Coverage		Docstring coverage is 3.45% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 29 functions across 5 files. (6 skipped: 6…	Write docstrings for the functions missing them to satisfy the coverage threshold.

Passed checks: 4 passed

Check name	Status	Explanation
Title check		The title clearly and concisely describes the main change: adding an optional local embedding server.
Description check		The description directly explains the new CLI command, OpenAI-compatible endpoint, local model support, optional dependencies, testing, and linked issue.
Linked Issues check		The PR implements the coding requirements in active issue #571. fastembed serve provides the requested CLI entry point and supports model selection, local model directories, device, host, port, and …
Out of Scope Changes check		The changes stay within issue #571. The workflow updates install the optional server dependencies for CI. The README and embedding server guide document the new service. The CLI entry point, dependenc…

Full details: Docstring Coverage

Explanation

Docstring coverage is 3.45% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 29 functions across 5 files. (6 skipped: 6 unsupported.)

• Fix all pre-merge checks with AI

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create a new PR

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}