chore(deps): Update graphrag-sdk requirement from <0.9.0,>=0.8.1 to >=0.8.1,<1.1.0

[dependabot]

Updates the requirements on graphrag-sdk to permit the latest version.

Release notes

Sourced from graphrag-sdk's releases.

v1.0.0

What's Changed

• docs: rc1 banner on README; fix CHANGELOG stable version by @​galshubeli in FalkorDB/GraphRAG-SDK#219
• fix(examples): guard against None score in PDF example context printing by @​galshubeli in FalkorDB/GraphRAG-SDK#221
• docs: README traffic refresh — vendor images, align embedder, add Mig… by @​galshubeli in FalkorDB/GraphRAG-SDK#220
• Revert "docs: README traffic refresh — vendor images, align embedder, add Mig…" by @​galshubeli in FalkorDB/GraphRAG-SDK#223
• Fix Cypher parse errors from control chars in GraphStore upserts by @​shahar-biron in FalkorDB/GraphRAG-SDK#222
• Docs/readme benchmark table by @​galshubeli in FalkorDB/GraphRAG-SDK#235
• refactor(schema): fold SchemaPattern into RelationType.patterns by @​galshubeli in FalkorDB/GraphRAG-SDK#236
• Docs/readme quickstart first by @​galshubeli in FalkorDB/GraphRAG-SDK#237
• chore(release): v1.0.0 by @​galshubeli in FalkorDB/GraphRAG-SDK#238
• docs: GraphRAG-Bench Novel benchmark refresh + 256-dim embedder default by @​galshubeli in FalkorDB/GraphRAG-SDK#239

New Contributors

• @​shahar-biron made their first contribution in FalkorDB/GraphRAG-SDK#222

Full Changelog: FalkorDB/GraphRAG-SDK@v1.0.0rc1...v1.0.0

Changelog

Sourced from graphrag-sdk's changelog.

[1.0.0] - 2026-04-21

First stable release of the v1.0 rewrite. pip install graphrag-sdk now resolves to this version by default. Legacy v0.x users can pin graphrag-sdk==0.8.2.

Added

• 9-step ingestion pipeline: Load, Chunk, Lexical Graph, Extract, Prune, Resolve, Write, Mentions (parallel), Index Chunks (parallel).
• Multi-path retrieval: Entity discovery via vector + fulltext search, relationship expansion, chunk retrieval, cosine reranking, and LLM-based answer generation.
• Strategy pattern: Swappable algorithms for every pipeline concern — chunking, extraction, resolution, retrieval, and reranking — each behind an abstract base class.
• GraphExtraction strategy: Two-step extraction using GLiNER2 for entity recognition (step 1) and LLM for relationship extraction and verification (step 2).
• Resolution strategies: ExactMatchResolution, DescriptionMergeResolution (LLM-assisted), SemanticResolution (embedding-based), and LLMVerifiedResolution.
• LiteLLM provider: Supports Azure OpenAI, OpenAI, Anthropic, Cohere, and 100+ LLM providers.
• OpenRouter provider: Alternative LLM/embedder provider via OpenRouter API.
• PDF ingestion: PdfLoader for processing PDF documents.
• Entity deduplication: finalize() post-ingestion step for dedup, embedding backfill, and index creation.
• Circuit breaker: Resilient FalkorDB connection with automatic failure detection and recovery.
• Multi-tenant support: Context with tenant isolation, distributed tracing, and latency budgeting.
• Parallel multi-source ingestion: ingest() accepts str | list[str] with max_concurrent parameter for bounded parallel ingestion.
• Retrieve/completion split: retrieve() for retrieval-only (no LLM call); completion() for full RAG pipeline with conversation history support.
• Native multi-turn conversations: completion(history=[...]) passes messages natively to the LLM provider's chat API (not string-stuffed into a single prompt). History accepts ChatMessage objects or {"role": ..., "content": ...} dicts with validated roles (system, user, assistant).
• ChatMessage model: Pydantic-validated message type with role: Literal["system", "user", "assistant"] and content: str. Exported from the top-level package.
• LLMInterface.ainvoke_messages(): New method for multi-turn message-based LLM calls. Default implementation falls back to ainvoke() (string concatenation), so custom providers work without changes. LiteLLM and OpenRouterLLM override with native implementations.
• Graph config node: __GraphRAGConfig__ singleton stores the embedding model and dimension used to build the graph; mismatches are caught on retrieval.
• Embedder.model_name: Abstract property on the Embedder ABC for identifying the embedding model.
• 556 tests: Comprehensive unit and integration test suite with mock providers.
• Full documentation: Architecture, strategies, configuration, providers, benchmark, and API reference.
• 4 examples: Quickstart, PDF with schema, custom strategies, and custom provider.

Changed

• query() is deprecated — use retrieve() for retrieval-only or completion() for full RAG.
• query_sync() is deprecated — use retrieve_sync() or completion_sync().
• ConnectionConfig.password field is hidden from repr() output.
• Dependency version bounds tightened: numpy<3, scipy<2, falkordb<2, gliner<1.
• pypdf minimum bumped to >=6.9.2 (CVE fixes).
• Development status classifier changed from Alpha to Production/Stable.

Fixed

• hnswlib import guard in SemanticResolution and LLMVerifiedResolution — raises clear ImportError instead of AttributeError when hnswlib is not installed.
• 14 ruff lint errors (import sorting, line length) resolved; CI no longer ignores lint rules.

Commits

• 87b54e7 Merge pull request #239 from FalkorDB/docs/salvage-from-224
• 15e7e31 Merge pull request #238 from FalkorDB/release/1.0.0
• 54e34d4 docs: update benchmark references to GraphRAG-Bench Novel + use text-embeddin...
• 3a6e618 chore(release): v1.0.0
• ec2c259 Merge pull request #237 from FalkorDB/docs/readme-quickstart-first
• 236ad61 docs(readme): replace rc1 pre-release banner with legacy v0.x install note
• 1e951b9 docs(readme): add Development Milestones section
• a331517 docs(readme): move Quick Start above pipeline description
• 5b1ddee Merge pull request #236 from FalkorDB/refactor/schema-patterns
• 56131c4 refactor(schema): trim RelationType — drop unused properties field, shrink pr...
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)