feat(memory-lancedb): Custom OpenAI BaseURL & Dimensions Support by rish2jain · Pull Request #17874 · openclaw/openclaw

rish2jain · 2026-02-16T08:02:51Z

Summary

This PR enables to support custom OpenAI-compatible endpoints (e.g. Ollama, vLLM) by allowing a configurable and custom vector dimensions.

Changes

Updated to include optional and .
Modified to skip dimension lookup if dimensions are explicitly provided.
Updated class to accept and use .
Relaxed model validation to allow custom models when dimensions are specified.

Motivation

Previously, the plugin hardcoded the OpenAI client initialization and model dimension lookup, preventing usage with local LLMs or other providers. This change makes the plugin provider-agnostic while maintaining default behavior for official OpenAI models.

Greptile Summary

Added support for custom OpenAI-compatible embedding providers (Ollama, vLLM, etc.) by introducing configurable baseUrl and dimensions parameters to the memory-lancedb plugin. The implementation maintains backward compatibility by keeping default OpenAI behavior when these optional fields are not provided.

Made embedding.model required (was optional) with a default fallback to text-embedding-3-small
Added optional embedding.baseUrl to customize the API endpoint
Added optional embedding.dimensions to specify vector dimensions for non-standard models
Modified dimension validation to skip lookup when dimensions are explicitly provided
Updated Embeddings class constructor to accept and use custom baseUrl
Relaxed model validation in JSON schema (removed enum constraint)

Confidence Score: 4/5

This PR is safe to merge with minor considerations around error handling.
The implementation is well-structured and maintains backward compatibility. The changes follow existing patterns in the codebase, properly validate inputs, and don't introduce security vulnerabilities. Score is 4 (not 5) because there's no test coverage for the new functionality.
No files require special attention

_{Last reviewed commit: c583a9a}

_{(3/5) Reply to the agent's comments like "Can you suggest a fix for this @greptileai?" or ask follow-up questions!}

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: e5fb52a814

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

extensions/memory-lancedb/config.ts

openclaw-barnacle · 2026-02-22T04:05:10Z

This pull request has been automatically marked as stale due to inactivity.
Please add updates or it will be closed.

greptile-apps

_{3 files reviewed, 1 comment}

_{Edit Code Review Agent Settings | Greptile}

greptile-apps · 2026-02-24T11:59:15Z

Additional Comments (1)

extensions/memory-lancedb/index.test.ts
Consider adding test coverage for the new baseUrl and dimensions configuration options to verify they're correctly parsed and passed through to the Embeddings class.

_{Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!}

Prompt To Fix With AI

This is a comment left during a code review.
Path: extensions/memory-lancedb/index.test.ts
Line: 66

Comment:
Consider adding test coverage for the new `baseUrl` and `dimensions` configuration options to verify they're correctly parsed and passed through to the Embeddings class.

<sub>Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!</sub>

How can I resolve this? If you propose a fix, please make it concise.

vincentkoc · 2026-02-27T05:13:16Z

Closing this in favour of #28318

vincentkoc · 2026-02-27T15:52:20Z

Deep review completed for #17874 with targeted cross-check against #20771 (closed as duplicate) and #17030 (overlapping).

Integrated improvements now committed locally:

Config hardening in extensions/memory-lancedb/config.ts
- Added embedding.baseUrl and embedding.dimensions parsing and validation.
- Improved unknown-model error to explicitly require either known models or embedding.dimensions.
- Added built-in dimensions for local/common models: nomic-embed-text (768), mxbai-embed-large (1024).
- Ensured OpenAI-default usage still requires an API key.
- Ensured local/custom endpoint usage can work without user-supplied key by assigning a safe local sentinel key ("local") for OpenAI SDK compatibility.
Runtime wiring in extensions/memory-lancedb/index.ts
- Passed optional baseUrl into OpenAI client initialization.
- Wired optional dimensions into vector dimension resolution.
Schema/UI sync in extensions/memory-lancedb/openclaw.plugin.json
- Added embedding.baseUrl and embedding.dimensions to schema and UI hints.
- Relaxed model enum to support OpenAI-compatible custom model names.
Test coverage in extensions/memory-lancedb/index.test.ts
- Added coverage for custom baseUrl + dimensions.
- Added known-model-without-dimensions coverage.
- Added unknown-model-without-dimensions failure coverage.
- Added local-endpoint-without-user-apiKey coverage.

Important review note:
I verified SDK behavior directly: openai constructor rejects missing apiKey even with custom baseURL. The parser-level local sentinel ("local") addresses this compatibility edge case while preserving no-user-key local endpoint UX.

Validation run:

pnpm test -- extensions/memory-lancedb/index.test.ts ✅ (15 passed, 1 skipped)
Full pnpm check currently fails in this workspace due to unrelated missing OTEL modules in extensions/diagnostics-otel (TS2307), outside this PR scope.

Related references:

* Changelog: add LanceDB custom baseUrl + dimensions entry (openclaw#17874) * Changelog: add Ollama autodiscovery hardening entry (openclaw#29201) * Changelog: add Ollama context-window unification entry (openclaw#29205) * Changelog: add compaction audit injection removal entry (openclaw#28507) * Changelog: add browser url alias entry (openclaw#29260) * Changelog: add codex weekly usage label entry (openclaw#26267)

…nclaw#17874) * feat(memory-lancedb): add custom baseUrl and dimensions support * fix(memory-lancedb): strict model typing and safe dimension resolution * style: fix formatting in memory-lancedb config * fix(memory-lancedb): sync manifest schema with new embedding options --------- Co-authored-by: OpenClaw Bot <bot@openclaw.ai>

* Changelog: add LanceDB custom baseUrl + dimensions entry (openclaw#17874) * Changelog: add Ollama autodiscovery hardening entry (openclaw#29201) * Changelog: add Ollama context-window unification entry (openclaw#29205) * Changelog: add compaction audit injection removal entry (openclaw#28507) * Changelog: add browser url alias entry (openclaw#29260) * Changelog: add codex weekly usage label entry (openclaw#26267)

…nclaw#17874) * feat(memory-lancedb): add custom baseUrl and dimensions support * fix(memory-lancedb): strict model typing and safe dimension resolution * style: fix formatting in memory-lancedb config * fix(memory-lancedb): sync manifest schema with new embedding options --------- Co-authored-by: OpenClaw Bot <bot@openclaw.ai>

* Changelog: add LanceDB custom baseUrl + dimensions entry (openclaw#17874) * Changelog: add Ollama autodiscovery hardening entry (openclaw#29201) * Changelog: add Ollama context-window unification entry (openclaw#29205) * Changelog: add compaction audit injection removal entry (openclaw#28507) * Changelog: add browser url alias entry (openclaw#29260) * Changelog: add codex weekly usage label entry (openclaw#26267)

…nclaw#17874) * feat(memory-lancedb): add custom baseUrl and dimensions support * fix(memory-lancedb): strict model typing and safe dimension resolution * style: fix formatting in memory-lancedb config * fix(memory-lancedb): sync manifest schema with new embedding options --------- Co-authored-by: OpenClaw Bot <bot@openclaw.ai>

* Changelog: add LanceDB custom baseUrl + dimensions entry (openclaw#17874) * Changelog: add Ollama autodiscovery hardening entry (openclaw#29201) * Changelog: add Ollama context-window unification entry (openclaw#29205) * Changelog: add compaction audit injection removal entry (openclaw#28507) * Changelog: add browser url alias entry (openclaw#29260) * Changelog: add codex weekly usage label entry (openclaw#26267)

…nclaw#17874) * feat(memory-lancedb): add custom baseUrl and dimensions support * fix(memory-lancedb): strict model typing and safe dimension resolution * style: fix formatting in memory-lancedb config * fix(memory-lancedb): sync manifest schema with new embedding options --------- Co-authored-by: OpenClaw Bot <bot@openclaw.ai>

* Changelog: add LanceDB custom baseUrl + dimensions entry (openclaw#17874) * Changelog: add Ollama autodiscovery hardening entry (openclaw#29201) * Changelog: add Ollama context-window unification entry (openclaw#29205) * Changelog: add compaction audit injection removal entry (openclaw#28507) * Changelog: add browser url alias entry (openclaw#29260) * Changelog: add codex weekly usage label entry (openclaw#26267)

* Changelog: add LanceDB custom baseUrl + dimensions entry (openclaw#17874) * Changelog: add Ollama autodiscovery hardening entry (openclaw#29201) * Changelog: add Ollama context-window unification entry (openclaw#29205) * Changelog: add compaction audit injection removal entry (openclaw#28507) * Changelog: add browser url alias entry (openclaw#29260) * Changelog: add codex weekly usage label entry (openclaw#26267) (cherry picked from commit 8090cb4) # Conflicts: # CHANGELOG.md