feat: allow configurable embedding model for local provider#55
Closed
eugenepro2 wants to merge 2 commits intotirth8205:mainfrom
Closed
feat: allow configurable embedding model for local provider#55eugenepro2 wants to merge 2 commits intotirth8205:mainfrom
eugenepro2 wants to merge 2 commits intotirth8205:mainfrom
Conversation
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
tirth8205
added a commit
that referenced
this pull request
Mar 26, 2026
Adds CRG_EMBEDDING_MODEL env var and model parameter to embedding functions, allowing users to specify any sentence-transformers compatible model. Changing the model re-embeds all nodes automatically. Co-Authored-By: eugenepro2 <noreply@github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2 tasks
tirth8205
added a commit
that referenced
this pull request
Mar 26, 2026
…nfigurable embeddings, MiniMax, Perl) * feat: integrate PR #43 — R language support Adds R language parsing with function extraction (both <- and = assignment), S4/R5 class detection via setClass/setRefClass, library/require/source imports, namespace-qualified calls (dplyr::filter), and testthat test detection. Co-Authored-By: michael-denyer <noreply@github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: integrate PR #54 — Vitest/Jest test detection Adds describe/it/test block parsing for JS/TS test files, producing synthetic Test nodes with description labels. Supports modifier suffixes (describe.only, it.skip, test.each) and nested describe/it containment. Co-Authored-By: JF10R <noreply@github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: integrate PR #53 — tsconfig path alias resolution Adds TsconfigResolver module that resolves TypeScript path aliases (e.g., @/ -> src/) from tsconfig.json compilerOptions.paths. Also resolves import targets to absolute file paths in IMPORTS_FROM edges. Co-Authored-By: JF10R <noreply@github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: integrate PR #58 — .mjs/.astro support Adds .mjs extension mapping to JavaScript and .astro extension mapping to TypeScript for import path resolution. Co-Authored-By: zoneghost7 <noreply@github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: integrate PR #55 — configurable embedding model Adds CRG_EMBEDDING_MODEL env var and model parameter to embedding functions, allowing users to specify any sentence-transformers compatible model. Changing the model re-embeds all nodes automatically. Co-Authored-By: eugenepro2 <noreply@github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: integrate PR #45 — MiniMax embedding provider Adds MiniMaxEmbeddingProvider using the embo-01 model (1536 dimensions) with support for distinct task types (db/query), batching, and retry logic. Co-Authored-By: octo-patch <noreply@github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat: integrate PR #62 — Perl support Adds Perl language parsing with package detection, subroutine extraction, use/require imports, and function call tracking. Includes test fixture and comprehensive test class. Co-Authored-By: potatogim <noreply@github.com> Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: michael-denyer <noreply@github.com> Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Owner
|
Integrated into main via PR #68. Thank you for the contribution! 🎉 |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
all-MiniLM-L6-v2modelparameter on MCP tool,CRG_EMBEDDING_MODELenv var, or falls back to the defaultChanges
embeddings.py:LocalEmbeddingProvideracceptsmodel_name;get_provider()andEmbeddingStoreforward newmodelkwarg; removed redundant default in Google provider passthroughtools.py:embed_graphandsemantic_search_nodesacceptmodelparameter — ensures query vectors use the same model as indexingmain.py: Both MCP tool wrappers (embed_graph_tool,semantic_search_nodes_tool) expose and forward themodelparameter; updated docstringsREADME.md: Configuration section documents custom model setup via.mcp.jsonenvtests/test_embeddings.py: +10 tests covering model_name priority chain, env var fallback, get_provider/EmbeddingStore passthrough, re-embed on provider changeConfiguration
Via
.mcp.json(recommended — applies to all MCP tool calls):{ "mcpServers": { "code-review-graph": { "command": "uvx", "args": ["code-review-graph", "serve"], "env": { "CRG_EMBEDDING_MODEL": "BAAI/bge-small-en-v1.5" } } } }Or directly via MCP tool parameter: