Release Notes

[2026-02-10]

llama-index-callbacks-wandb [0.4.2]

• Fix potential crashes and improve security defaults in core components (#20610)

llama-index-core [0.14.14]

• fix: catch pydantic ValidationError in VectorStoreQueryOutputParser (#20450)
• fix: distinguish empty string from None in MediaResource.hash (#20451)
• Langchain1.x support (#20472)
• Fix DeprecationWarning: 'asyncio.iscoroutinefunction' is deprecated (#20517)
• fix(core): fallback to bundled nltk cache if env var missing (#20528)
• feat(callbacks): add TokenBudgetHandler for cost governance (#20546)
• fix(core):handled a edge case in truncate_text function (#20551)
• fix(core):fix in types Thread passing None when target is None instead of copy_context().run (#20553)
• chore: bump llama-index lockfile, and minor test tweaks (#20556)
• Compatibility for workflows context changes (#20557)
• test(core): fix cache dir path test for Windows compatibility (#20566)
• fix(tests): enforce utf-8 encoding in json reader tests for windows compatibility (#20576)
• Fix BM25Retriever mapping in upgrade tool / 修复升级工具中的 BM25Retriever 映射 (#20582)
• fix(agent): handle empty LLM responses with retry logic and add test cases (#20596)
• fix: add show_progress parameter to run_transformations to prevent unexpected keyword argument error (#20608)
• Fix potential crashes and improve security defaults in core components (#20610)
• Add core 3.14 tests (#20619)

llama-index-embeddings-cohere [0.7.0]

• fix(embeddings-cohere): add retry logic with tenacity (#20592)

llama-index-embeddings-google-genai [0.3.2]

• Add client headers to Gemini API requests (#20519)

llama-index-embeddings-siliconflow [0.3.2]

• Fix DeprecationWarning: 'asyncio.iscoroutinefunction' is deprecated (#20517)

llama-index-embeddings-upstage [0.5.1]

• chore(deps): bump the uv group across 4 directories with 4 updates (#20531)

llama-index-graph-stores-falkordb [0.4.2]

• fix(falkordb): Fix MENTIONS relationship creation with triplet_source_id (#20650)

llama-index-llms-anthropic [0.10.8]

• chore: Update cacheable Anthropic models (#20581)
• chore: add support for opus 4.6 (#20635)

llama-index-llms-bedrock-converse [0.12.8]

• fix bedrock converse empty tool config issue (#20571)
• fix(llms-bedrock-converse): improve bedrock converse retry handling (#20590)
• feat(bedrock-converse): Add support for Claude Opus 4.6 (#20637)
• Add support for adaptive thinking in Bedrock (#20659)
• chore(deps): bump the pip group across 2 directories with 7 updates (#20662)

llama-index-llms-cohere [0.7.1]

• Feat: add custom base_url support to Cohere LLM (#20534)
• fix(llms-cohere): handle additional error types in retry logic (#20591)

llama-index-llms-dashscope [0.5.2]

• fix(dashscope): remove empty tool_calls from assistant messages (#20535)

llama-index-llms-google-genai [0.8.7]

• Add client headers to Gemini API requests (#20519)
• fix(decorator):adds logic to llm_retry_decorator for async methods. (#20588)
• Fix/google genai cleanup (#20607)
• fix(google-genai): skip model meta fetch when not needed (#20639)

llama-index-llms-huggingface-api [0.6.2]

• Update sensible default provider for huggingface inference api (#20589)

llama-index-llms-langchain [0.7.1]

• Langchain1.x support (#20472)

llama-index-llms-openai [0.6.18]

• OpenAI response fix (#20538)
• feat: Add support for gpt-5.2-chat model (#20549)
• fix(openai): make image_url detail optional in message dict (#20609)
• Add new reasoning types (#20612)
• fix(openai): exclude unsupported params for all reasoning models (#20627)

llama-index-llms-openai-like [0.6.0]

• make transformers an optional dependency for openai-like (#20580)

llama-index-llms-openrouter [0.4.4]

• make transformers an optional dependency for openai-like (#20580)

llama-index-llms-siliconflow [0.4.3]

• Fix DeprecationWarning: 'asyncio.iscoroutinefunction' is deprecated (#20517)

llama-index-llms-upstage [0.7.0]

• add new upstage model(solar-pro3) (#20544)

llama-index-llms-vllm [0.6.2]

• feat: add openai-like server mode for VllmServer (#20537)

llama-index-memory-bedrock-agentcore [0.1.2]

• Add event and memory record deletion methods in bedrock-agentcorememory (#20428)
• chore(deps): update llama-index-core dependency lock to include 0.14.x (#20483)

llama-index-memory-mem0 [1.0.0]

• fix: mem0 integration cleanup + refactor (#20532)

llama-index-node-parser-chonkie [0.1.1]

• feat: add chonkie integration (#20622)
• update readme (#20656)

llama-index-node-parser-docling [0.4.2]

• fix: catch pydantic ValidationError in VectorStoreQueryOutputParser (#20450)

llama-index-packs-code-hierarchy [0.6.1]

• chore(deps): bump the uv group across 12 directories with 14 updates (#20578)

llama-index-packs-gmail-openai-agent [0.4.1]

• chore(deps): bump the uv group across 12 directories with 14 updates (#20578)

llama-index-packs-multidoc-autoretrieval [0.4.1]

• chore(deps): bump the uv group across 12 directories with 14 updates (#20578)

llama-index-packs-panel-chatbot [0.4.1]

• chore(deps): bump the uv group across 12 directories with 14 updates (#20578)

llama-index-packs-recursive-retriever [0.7.1]

• chore(deps): bump the uv group across 12 directories with 14 updates (#20578)
• chore(deps): bump the pip group across 2 directories with 7 updates (#20662)

llama-index-packs-resume-screener [0.9.3]

• chore(deps): bump the uv group across 12 directories with 14 updates (#20578)

llama-index-packs-retry-engine-weaviate [0.5.1]

• chore(deps): bump the uv group across 12 directories with 14 updates (#20578)

llama-index-packs-streamlit-chatbot [0.5.2]

• chore(deps): bump the uv group across 12 directories with 14 updates (#20578)

llama-index-packs-sub-question-weaviate [0.4.1]

• chore(deps): bump the uv group across 12 directories with 14 updates (#20578)

llama-index-packs-timescale-vector-autoretrieval [0.4.1]

• chore(deps): bump the uv group across 12 directories with 14 updates (#20578)

llama-index-postprocessor-cohere-rerank [0.6.0]

• fix(cohere-rerank): add retry logic and tenacity dependency to cohere rerank (#20593)

llama-index-postprocessor-nvidia-rerank [0.5.4]

• fix(nvidia-rerank): fix initialization logic for on-prem auth (#20560)
• fix(nvidia-rerank): correct private attribute reference (#20570)
• fix(nvidia-rerank): Fix POST request url for locally hosted NIM rerankers (#20579)

llama-index-postprocessor-tei-rerank [0.4.2]

• fix(tei-rerank): use index field from API response for correct score … (#20599)
• test(tei-rerank): add test coverage for rerank retry coverage (#20600)

llama-index-protocols-ag-ui [0.2.4]

• fix: avoid ValueError in ag-ui message conversion for multi-block ChatMessages (#20648)

llama-index-readers-datasets [0.1.0]

• chore(deps): bump the uv group across 4 directories with 4 updates (#20531)

llama-index-readers-microsoft-sharepoint [0.7.0]

• Sharepoint page support events (#20572)

llama-index-readers-obsidian [0.6.1]

• Langchain1.x support (#20472)

llama-index-readers-service-now [0.2.2]

• chore(deps): bump the pip group across 2 directories with 7 updates (#20662)

llama-index-tools-mcp [0.4.6]

• feat: implement partial_params support to McpToolSpec (#20554)

llama-index-tools-mcp-discovery [0.1.0]

• Add llama-index-tools-mcp-discovery integration (#20502)

llama-index-tools-moss [0.1.0]

• feat(tools): add Moss search engine integration (#20615)

llama-index-tools-seltz [0.1.0]

• feat(tools): add Seltz web knowledge tool integration (#20626)

llama-index-tools-typecast [0.1.0]

• Migrate Typecast tool to V2 API for voices endpoints (#20548)

llama-index-tools-wolfram-alpha [0.5.0]

• feat(wolfram-alpha): switch to LLM API with bearer auth (#20586)

llama-index-vector-stores-clickhouse [0.6.2]

• fix(clickhouse): Add drop_existing_table parameter to prevent data loss (#20651)

llama-index-vector-stores-milvus [0.9.6]

• chore(deps): bump the uv group across 4 directories with 4 updates (#20531)

llama-index-vector-stores-mongodb [0.9.1]

• Update MongoDB vector store tests to use newer model (#20515)

llama-index-vector-stores-oceanbase [0.4.0]

• feat(oceanbase): add sparse/fulltext/hybrid search (#20524)

llama-index-vector-stores-opensearch [1.0.0]

• Changed OpenSearch engine default from deprecated nmslib to faiss (#20507)
• chore(deps): bump the uv group across 4 directories with 4 updates (#20531)

llama-index-vector-stores-postgres [0.7.3]

• fix(postgres): disable bitmap scan for vector queries (#20514)

llama-index-vector-stores-yugabytedb [0.5.4]

• Add YugabyteDB as a Vector Store (#20559)
• chore(deps): bump the pip group across 2 directories with 7 updates (#20662)

llama-index-voice-agents-gemini-live [0.2.2]

• Add client headers to Gemini API requests (#20519)