Skip to content

build(deps): Bump the python-dependencies group in /sdk/python with 4 updates - #752

Merged
NiveditJain merged 1 commit into
mainfrom
dependabot/uv/sdk/python/python-dependencies-5a83365e11
Aug 25, 2026
Merged

NiveditJain merged 1 commit into
mainfrom
dependabot/uv/sdk/python/python-dependencies-5a83365e11

Conversation

@dependabot

@dependabot dependabot Bot commented on behalf of github Aug 24, 2026 •

Copy link
Copy Markdown
Contributor

Bumps the python-dependencies group in /sdk/python with 4 updates: langchain-core, crewai, llama-index-core and pydantic-ai-slim.

Updates langchain-core from 1.5.6 to 1.6.0

Release notes

Sourced from langchain-core's releases.

langchain-core==1.6.0

Changes since langchain-core==1.5.6

release(core): 1.6.0 (#39760) fix(core): resolve postponed annotations in StructuredTool._injected_args_keys (#39602) feat(core): add standard model exception types (#39538) fix(core): allow deserializing RunnablePick (#39753) fix(core): make convert_to_openai_function handle callables and non-dict mappings (#39750) fix(core): make subprocess and temporary file tests portable on Windows (#39664) fix(core): fail fast when tool schemas can't resolve forward refs during serialization (#39570) test(core): avoid version-dependent runnable snapshots (#39705) fix(core): require all nested properties for strict tool schemas (#39306) fix(core): remove stale sync-stream xfail [closes #39720] (#39723) perf(core): Lazily import transformers (#38037) fix(core): accept non-dict Mapping values in mustache templates (#39680) docs(core): clarify Runnable pipe coercion [closes #39075] (#39707) fix(core): finalize chain-group runs on BaseException (#39699)

Commits
  • 85602c3 release(core): 1.6.0 (#39760)
  • 5c3538e fix(core): resolve postponed annotations in `StructuredTool._injected_args_ke...
  • 9984a87 feat(core): add standard model exception types (#39538)
  • b3e9eef chore(model-profiles): refresh model profile data (#39751)
  • ded2a1f fix(core): allow deserializing RunnablePick (#39753)
  • 04ae744 fix(core): make convert_to_openai_function handle callables and non-dict ma...
  • 37f2662 feat(langchain): support custom token_counter in ContextEditingMiddleware (#3...
  • 2019bf5 fix(openai): raise clear error on unexpected response type in `_create_chat_r...
  • e92c6db fix(langchain): re-raise non-retryable exceptions in ModelRetryMiddleware (#3...
  • f368888 fix(partners): isolate unit tests from network [closes #39727] (#39729)
  • Additional commits viewable in compare view

Updates crewai from 1.15.16 to 1.15.17

Release notes

Sourced from crewai's releases.

1.15.17

What's Changed

Features

  • Add declarative conversational flows documentation
  • Synthesize built-in conversational methods for declarations
  • Enable declarations to drive conversational mode
  • Make conversational opt-in unmistakable
  • Carry the AMP slug on tools resolved from a slug reference
  • Handle oversized single messages during chunking

Bug Fixes

  • Fix usage of the URL hostname as MCP HTTP and SSE server_name
  • Close the agent scope on every failed attempt
  • Attribute tool errors to the tool that failed
  • Pin SSRF checks to each redirect hop and peer IP
  • Resolve issues with native tool calls broken over OpenAI Responses API

Documentation

  • Update documentation with a snapshot and changelog for v1.15.16

Contributors

@​Copilot, @​Vidit-Ostwal, @​github-code-quality[bot], @​joaomdmoura, @​lorenzejay, @​lucasgomide, @​theCyberTech

Commits
  • b3ab193 [docs-freeze] docs: snapshot and changelog for v1.15.17 (#7055)
  • 5ddf62a feat: bump versions to 1.15.17 (#7054)
  • 4dfd074 docs(flow): document declarative conversational flows (#7035)
  • 01a156d feat(flow): synthesize the built-in conversational methods for declarations (...
  • 6f16e74 feat(flow): let a declaration drive conversational mode (#7032)
  • dabd123 fix: use the URL hostname as MCP HTTP and SSE server_name (#7048)
  • f0e00df feat(flow): make conversational opt-in unmistakable (#7031)
  • 47fa48d fix: close the agent scope on every failed attempt (#6997)
  • b1c9c89 fix(telemetry): attribute tool errors to the tool that failed (#7043)
  • 7942ce0 test(flow): unskip the conversational graph end-to-end suite (#7030)
  • Additional commits viewable in compare view

Updates llama-index-core from 0.14.23 to 0.14.24

Release notes

Sourced from llama-index-core's releases.

v0.14.24

Release Notes

[2026-08-19]

llama-index-core [0.14.24]

  • fix: close HotpotQA evaluator file handles (#22113)
  • fix(core): preserve oversized leaf nodes in CodeSplitter (#22119)
  • fix: preserve multiblock chat history writes (#22124)
  • fix(core): skip empty nodes in HTMLNodeParser (#22125)
  • fix(core): respect mmr_threshold=0 in MMR embedding search (#22126)
  • fix: keep all nodes per document in IngestionPipeline upserts (re-submit of #22046, not a dup of #21456) (#22133)
  • fix: don't mark *args/**kwargs as required tool parameters (#22135)
  • fix(core): default MetadataFilters condition to AND when None (#22155)
  • fix: honor agent structured_output_fn/output_cls inside AgentWorkflow (#22162)
  • fix(core): use word tokenization for stopword removal in SemanticDoubleMergingSplitterNodeParser (#22167)
  • fix(core): restore prompt helper fallback to supplied LLM metadata (#22177)
  • populate response text on streaming chat response when writing to memory (#22179)
  • feat(property_graph): add raise_on_error to LLM path extractors (#22195)
  • Clarify FactExtractionMemoryBlock condense prompt to return a full snapshot (#22213)
  • fix a few typos (#22512)
  • fix(core): give each CitationQueryEngine citation node its own id and offsets (#22537)
  • fix(core): persist SimpleChatStore without escaping non-ascii (#22538)
  • feat(core): allow Memory to accept any AsyncDBChatStore (#22541)
  • test(node_parser): enable pytest discovery for sentence_window tests (#22545)
  • Implement async for LLMRerank (#22597)

llama-index-embeddings-bedrock [0.8.3]

  • fix: make aioboto3 optional in llama-index-embeddings-bedrock (#21915)
  • fix(embeddings): validate non-empty Bedrock embedding payload before AWS call (#22527)

llama-index-embeddings-voyageai [0.6.1]

  • fix: set chunk_size=32000 and fix input format for contextualizedembeddings (#22144)

llama-index-graph-stores-falkordb [0.6.0]

  • fix(falkordb): correct schema, vector ranking, filters and batching in property graph store (#22522)

llama-index-indices-managed-dashscope [0.5.1]

  • fix(dashscope): send query_history in retriever payload (#22212)

llama-index-indices-managed-lancedb [0.3.2]

  • fix(lancedb): repair image ingestion crash, embedding_model typo, and… (#22544)
  • lint (#22572)

... (truncated)

Changelog

Sourced from llama-index-core's changelog.

llama-index-core [0.14.24]

  • fix: close HotpotQA evaluator file handles (#22113)
  • fix(core): preserve oversized leaf nodes in CodeSplitter (#22119)
  • fix: preserve multiblock chat history writes (#22124)
  • fix(core): skip empty nodes in HTMLNodeParser (#22125)
  • fix(core): respect mmr_threshold=0 in MMR embedding search (#22126)
  • fix: keep all nodes per document in IngestionPipeline upserts (re-submit of #22046, not a dup of #21456) (#22133)
  • fix: don't mark *args/**kwargs as required tool parameters (#22135)
  • fix(core): default MetadataFilters condition to AND when None (#22155)
  • fix: honor agent structured_output_fn/output_cls inside AgentWorkflow (#22162)
  • fix(core): use word tokenization for stopword removal in SemanticDoubleMergingSplitterNodeParser (#22167)
  • fix(core): restore prompt helper fallback to supplied LLM metadata (#22177)
  • populate response text on streaming chat response when writing to memory (#22179)
  • feat(property_graph): add raise_on_error to LLM path extractors (#22195)
  • Clarify FactExtractionMemoryBlock condense prompt to return a full snapshot (#22213)
  • fix a few typos (#22512)
  • fix(core): give each CitationQueryEngine citation node its own id and offsets (#22537)
  • fix(core): persist SimpleChatStore without escaping non-ascii (#22538)
  • feat(core): allow Memory to accept any AsyncDBChatStore (#22541)
  • test(node_parser): enable pytest discovery for sentence_window tests (#22545)
  • Implement async for LLMRerank (#22597)

llama-index-embeddings-bedrock [0.8.3]

  • fix: make aioboto3 optional in llama-index-embeddings-bedrock (#21915)
  • fix(embeddings): validate non-empty Bedrock embedding payload before AWS call (#22527)

llama-index-embeddings-voyageai [0.6.1]

  • fix: set chunk_size=32000 and fix input format for contextualizedembeddings (#22144)

llama-index-graph-stores-falkordb [0.6.0]

  • fix(falkordb): correct schema, vector ranking, filters and batching in property graph store (#22522)

llama-index-indices-managed-dashscope [0.5.1]

  • fix(dashscope): send query_history in retriever payload (#22212)

llama-index-indices-managed-lancedb [0.3.2]

  • fix(lancedb): repair image ingestion crash, embedding_model typo, and… (#22544)
  • lint (#22572)

llama-index-llms-anthropic [0.11.10]

  • feat(anthropic): Add support for Claude Sonnet 5 (#22202)
  • Fix/sonnet 5 function calling (#22217)
  • feat(anthropic, bedrock-converse): add Claude Opus 5 to model allowlists (#22451)

... (truncated)

Commits
  • 9ba74b8 Release 0.14.24 (#22761)
  • 04823eb fix pre-release script (#22760)
  • 30e759a feat(google-genai): use gemini 3.7 Flash by default in lib and docs (#22734)
  • 01cb24c fix: dont overwrite thought signatures (#21562)
  • 3759b52 feat(ag-ui): support multimodal user input (images, audio, video, documents) ...
  • afd0fef fix(pinecone): support pinecone client v8 and v9 (#22511)
  • 9bbdfe1 fix a few typos (#22512)
  • 9808e4b fix(embeddings): validate non-empty Bedrock embedding payload before AWS call...
  • 7b64c7e fix: close file handles and http connections in reader integrations (#22529)
  • 8c983cf fix: don't hardcode verify=False for S3 TLS in BotoMinioReader (#22530)
  • Additional commits viewable in compare view

Updates pydantic-ai-slim from 2.32.0 to 2.33.0

Release notes

Sourced from pydantic-ai-slim's releases.

v2.33.0 (2026-08-20)

⚠️ If Anthropic stopped working for you in the last day

anthropic 1.0.0 — rebuilt on httpx2, with legacy httpx support removed — reached PyPI on August 20. Every earlier pydantic-ai release, including v2.32.2 cut a few hours before this one, allowed that version without supporting it: a fresh or unpinned install of pydantic-ai[anthropic] could resolve anthropic 1.0.0 and then fail at runtime when using an Anthropic model. Apologies for the breakage window.

  • Fix: upgrade to pydantic-ai v2.33.0, which requires and supports anthropic>=1.0.0.
  • If you need to stay on an older pydantic-ai: pin anthropic<1.

If you pass your own http_client to AnthropicProvider, it must now be an httpx2.AsyncClient — the 1.x SDK rejects legacy httpx clients at construction.

What's Changed

⚠️ Compatibility Notes

Full Changelog: pydantic/pydantic-ai@v2.32.2...v2.33.0

v2.32.2 (2026-08-20)

What's Changed

🐛 Bug Fixes

New Contributors

Full Changelog: pydantic/pydantic-ai@v2.32.1...v2.32.2

v2.32.1 (2026-08-19)

What's Changed

🐛 Bug Fixes

Full Changelog: pydantic/pydantic-ai@v2.32.0...v2.32.1

Commits
  • 1d7eb69 Use httpx2 for Anthropic clients (#7657)
  • 5e59fef Fail the workflow on UnexpectedModelBehavior and FallbackExceptionGroup u...
  • 9226bfd docs: deep-link the Logfire LLM-observability and evals pages (#7656)
  • 8a34441 Guard released API compatibility in CI (#7652)
  • 8296f26 Make documentation prose more concrete (#7644)
  • e5b2aa9 Normalize DeepSeek Responses function-call replay (#7448)
  • f66c07b Recognize m.youtube.com URLs in VideoUrl (#7592)
  • 07cb795 Fix dead link to xAI voice-agent guide in docs/realtime/xai.md (#7503)
  • 53aad1d Fix RunContext.cancel() in realtime sessions (#7528)
  • 57687af Await async callable instances in pydantic_evals tasks and the evaluate d...
  • 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 <dependency name> major version will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
  • @dependabot ignore <dependency name> minor version will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
  • @dependabot ignore <dependency name> will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
  • @dependabot unignore <dependency name> will remove all of the ignore conditions of the specified dependency
  • @dependabot unignore <dependency name> <ignore condition> will remove the ignore condition of the specified dependency and ignore conditions

Bumps the python-dependencies group in /sdk/python with 4 updates: [langchain-core](https://github.com/langchain-ai/langchain), [crewai](https://github.com/crewAIInc/crewAI), [llama-index-core](https://github.com/run-llama/llama_index) and [pydantic-ai-slim](https://github.com/pydantic/pydantic-ai).


Updates `langchain-core` from 1.5.6 to 1.6.0
- [Release notes](https://github.com/langchain-ai/langchain/releases)
- [Commits](langchain-ai/langchain@langchain-core==1.5.6...langchain-core==1.6.0)

Updates `crewai` from 1.15.16 to 1.15.17
- [Release notes](https://github.com/crewAIInc/crewAI/releases)
- [Commits](crewAIInc/crewAI@1.15.16...1.15.17)

Updates `llama-index-core` from 0.14.23 to 0.14.24
- [Release notes](https://github.com/run-llama/llama_index/releases)
- [Changelog](https://github.com/run-llama/llama_index/blob/main/CHANGELOG.md)
- [Commits](run-llama/llama_index@v0.14.23...v0.14.24)

Updates `pydantic-ai-slim` from 2.32.0 to 2.33.0
- [Release notes](https://github.com/pydantic/pydantic-ai/releases)
- [Changelog](https://github.com/pydantic/pydantic-ai/blob/main/docs/changelog.md)
- [Commits](pydantic/pydantic-ai@v2.32.0...v2.33.0)

---
updated-dependencies:
- dependency-name: langchain-core
  dependency-version: 1.6.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: python-dependencies
- dependency-name: crewai
  dependency-version: 1.15.17
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: python-dependencies
- dependency-name: llama-index-core
  dependency-version: 0.14.24
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: python-dependencies
- dependency-name: pydantic-ai-slim
  dependency-version: 2.33.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: python-dependencies
...

Signed-off-by: dependabot[bot] <support@github.com>
@dependabot dependabot Bot added dependencies Pull requests that update a dependency file python:uv Pull requests that update python:uv code labels Aug 24, 2026
@github-actions

Copy link
Copy Markdown
Contributor

Thanks @dependabot[bot] for keeping our dependencies current! 🙌

@github-actions github-actions Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Automated approval for Dependabot dependency update.

@hermes-exosphere hermes-exosphere left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Approved automatically by the Hermes Dependabot approver.

@NiveditJain
NiveditJain merged commit 80bb3ce into main Aug 25, 2026
23 checks passed
@dependabot
dependabot Bot deleted the dependabot/uv/sdk/python/python-dependencies-5a83365e11 branch August 25, 2026 12:09
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

dependencies Pull requests that update a dependency file python:uv Pull requests that update python:uv code

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants