v1.0.6

What's Changed

• chore: update llm.py by @eltociear in #436
• fix(lightrag): Fix the logic for handling chunks with entities but no relationships by @magicyuan876 in #443
• Fix/lazy import by @davidleon in #444
• fix: rare embedding issue. by @davidleon in #440
• Feature/add concurrent embedding limit by @davidleon in #441
• Fix/extra keyword extraction by @davidleon in #442
• feat: Add ChromaDB integration for vector storage by @spo0nman in #447
• support TiDB: add TiDBKVStorage, TiDBVectorDBStorage by @Weaxs in #452
• fix: fix variable name(entitiy->entity) by @ChenZiHong-Gavin in #458
• fix: update operate.py by @tjyiiuan in #461
• Maybe very important!!! Fix embedding error by @billvsme in #464
• Supporting Zhipu AI API by @JasonGuoo in #462

New Contributors

• @eltociear made their first contribution in #436
• @spo0nman made their first contribution in #447
• @Weaxs made their first contribution in #452
• @tjyiiuan made their first contribution in #461
• @billvsme made their first contribution in #464
• @JasonGuoo made their first contribution in #462

Full Changelog: v1.0.5...v1.0.6

v1.0.5

What's Changed

• Added the ability to use embedding cache during queries by @magicyuan876 in #406
• Fixed a bug where args_hash was not calculated during embedding caching due to only being computed in regular caching scenarios by @magicyuan876 in #408
• Update oracle_impl.py by @zsuroy in #409
• Add support for Ollama streaming output and integrate Open-WebUI as the chat UI demo by @partoneplay in #407
• Refactored caching logic by @magicyuan876 in #412
• Optimized embedding similarity caching mechanism, added LLM similarity check functionality, and improved caching mechanism by @magicyuan876 in #421
• Added Jina embedding support by @davidleon in #423
• Interactive Graph: Mouse hover on nodes and edges displays description in a pop-up window by @kaushikacharya in #424
• Fixed a bug #306 by @magicyuan876 in #428
• Fix: Ensure entity_or_relation_name is a string in _handle_entity_relation_summary by @SaujanyaV in #415
• Add an Ollama API service demo by @zhenya-zhu in #430
• Add support for OpenAI-compatible streaming output and delete unreachable code by @partoneplay in #417
• Fix: SyntaxError by @guyuecode in #425
• Fix extra kwargs error: keyword_extraction by @davidleon in #429
• Fix (LightRAG): Use is_closed() instead of _closed by @ChenZiHong-Gavin in #432
• Fix Unicode escape for cases like "/utils" in the response by @davidleon in #435
• Optimized keyword extraction prompt by @magicyuan876 in #434

New Contributors

• @zsuroy made their first contribution in #409
• @kaushikacharya made their first contribution in #424
• @SaujanyaV made their first contribution in #415
• @zhenya-zhu made their first contribution in #430
• @guyuecode made their first contribution in #425
• @ChenZiHong-Gavin made their first contribution in #432

Full Changelog: v1.0.4...v1.0.5

v1.0.4

What's Changed

• Add MongoDB as KV storage by @partoneplay in #399
• Fix JSON parsing error

Full Changelog: v1.0.3...v1.0.4

v1.0.3

Fix: unexpected keyword argument error.