- ✨ Integrate a tool for the naver_search - PR #30392
- ✨ Add configurable text key for indexing and the retriever in Pinecone Hybrid Search - PR #29697
- 📝 Modify parameter for the LoRA adapter on the vllm page - PR #27930
- 📝 Add content for the Lora adapter in the VLLM page - PR #27788
- ✨ Add vLLM integration to apply lora_request - PR #27731
A python package that integrates Naver's various APIs with LangChain, making it easier to build powerful applications using Naver's services within the LangChain ecosystem.
- 🐛 remove unused parameter:
output_type- PR #33
- 🐛 same Items in parsed_result by Concatenating Each Parsed Result - PR #30392
- ✨ Add dynamic embedding models - PR #1063
- ✨ Add intfloat/multilingual-e5-large-instruct to embedding model - PR #1061
- 🐛 Fix ID Mismatch Error in VectorDB During Evaluation - PR #1056
- 📝 Update Milvus configuration examples - PR #1030
- 📝 Add warning about AttributeError: vllm_model - PR #1014
- 🐛 Refactor method to properly release vllm instance resources - PR #1012
- ✨ Add a function to generate multiple quries - PR #1009
- ✨ Set parameters for Milvus using the configuration file - PR #1001
- 🐛 Change default similarity_metric for milvus - PR #1000
- 📝 Modify contents on upstage parser - PR #967
- 🐛 Add explicit error handling and improve diagnostics for uninitialized process group access in
_world.pg_map- PR #141796
- ✨ Add support for Sub-Issues - PR #3580
- ✨ Add support for Sub-Issues - PR #3258
- ✨ Add KiwiTokenizer for Korean text tokenization - PR #70