Python Releases
Python language news, releases and ecosystem updates. New features, breaking changes, security advisories and deprecations - each summarised in plain English and updated continuously.
Tracking 132 Python releases · Updated
- CrewAI Releases sdkaiengineerpython-agentic ·
crewAI 1.15.10 Patch Release
This patch release for crewAI introduces a new feature for collecting skill usage events, enhancing observability into agent operations. It also includes several documentation updates, such as removing migrated AMP documentation and updating security reporting guidelines. The changes are aimed at developers and architects using crewAI for agent orchestration.
feature patch - Google ADK (Python) Releases sdkaiengineermediaenergypython-agentic ·
Google ADK Python v2.6.0 adds features, fixes
Google's ADK Python library has been updated to version 2.6.0, introducing numerous new features and improvements. Key additions include enhanced support for authentication, new evaluation samples for agents, expanded tooling for Eventarc and BigQuery, and improved telemetry and CLI capabilities. These changes aim to provide developers with more robust tools for building and deploying AI agents, with a particular focus on enhanced flexibility and integration within the Google ecosystem.
feature patch - CrewAI Releases sdkaiengineerpython-agentic ·
crewAI 1.15.9: Improved Tool Failure Reporting and Flow Events
CrewAI version 1.15.9 enhances the platform by surfacing tool failures accurately instead of misreporting them as successes. It also introduces a new FlowFailedEvent to signal execution failures and implements progressive disclosure for skills. These changes are relevant for developers and architects using CrewAI for agent orchestration, providing clearer insights into execution issues.
feature patch - Pydantic AI Releases sdkaiengineerpython-agentic ·
pydantic-ai v2.21.0 adds token limit and refreshes model names
Pydantic-ai v2.21.0 introduces a `per_request_input_tokens_limit` parameter to `UsageLimits`, allowing finer control over request processing. The release also includes a bug fix to refresh `KnownModelName` using Gateway probes and stable Google image IDs. These updates are now available and primarily benefit developers building AI applications with pydantic-ai.
feature patch - OpenAI Agents SDK (Python) Releases sdkaiengineerpython-agentic ·
OpenAI Agents Python v0.19.1: Bug fixes and documentation updates
This release of the OpenAI Agents Python library addresses multiple bugs and includes documentation improvements. Key fixes involve enhancing sandbox path grants, redacting sensitive information in logs, and improving WebSocket error handling and streaming model behavior. These changes benefit developers using the library by increasing stability and reliability, particularly for complex agent interactions and data processing tasks.
patch - Pydantic AI Releases sdkaipython-agentic ·
pydantic-ai v2.20.0: New Claude Opus, OpenAI Responses API, and Usage Fixes
This release of pydantic-ai introduces support for Claude Opus 5 and enhances the OpenAI Responses API with reasoning context. It also includes several bug fixes, such as preserving arbitrary usage fields, capturing Anthropic thinking tokens, and correcting handling for embeddings and tool-search history. These updates are beneficial for developers integrating AI models, particularly those using Anthropic, OpenAI, and Vertex AI.
feature patch - LangGraph Releases sdkaiengineerpython-agentic ·
LangGraph 1.2.10: Stream events typing, trace policy changes
LangGraph version 1.2.10 introduces type hinting for `stream_events` return and native projections, enhancing developer experience. It also includes changes to the trace policy, removing the `TracePolicy` and exposing `trace_policy` on `add_node`. The update affects developers using LangGraph for building complex, stateful applications and includes several dependency updates.
patch feature - CrewAI Releases sdkaiengineerpython-agentic ·
crewAI 1.15.8: WaitTool, FileWriter fix, and docs
Version 1.15.8 of crewAI introduces a new WaitTool for pausing long-running jobs, enhancing agent control. It also fixes issues with the FileWriterTool and marks E2B_API_KEY as a required environment variable. These updates are primarily for developers and engineers using crewAI for agent orchestration.
feature patch - Pydantic AI Releases sdkaiengineerpython-agentic ·
Pydantic-AI v2.19.0 adds headers and retry_after to ModelHTTPError
Pydantic-AI version 2.19.0 introduces new fields, `headers` and `retry_after`, to the `ModelHTTPError` class, making error handling more informative by capturing details from various provider SDKs. This release also includes a bug fix related to external cancellation assertion and updates a dependency. These changes are aimed at improving robustness for developers integrating AI models with Pydantic.
feature patch - OpenAI Agents SDK (Python) Releases sdkaimediapython-agentic ·
OpenAI Agents Python: Programmatic Tool Calling and SDK Enhancements
This release introduces programmatic tool calling, allowing OpenAI models to generate JavaScript for tool coordination. The update also standardizes SDK configuration, enhances error logging for security, and improves compatibility with various LLM providers. These changes benefit developers integrating OpenAI agents into applications, providing more control and robustness.
feature patch - Pydantic AI Releases sdkaiawsengineerpython-agentic ·
pydantic-ai v2.18.0 adds Anthropic/OpenRouter tool support, web access
Version 2.18.0 of pydantic-ai introduces new features for integrating with AI services. It adds AdvisorTool support for Anthropic and OpenRouter, enabling more flexible AI agent development. Additionally, it includes external web access options for the WebSearchTool with OpenAI, and improves multi-region support for Google Cloud Provider. These updates are primarily relevant for developers building AI-powered applications with pydantic-ai.
feature patch - CrewAI Releases sdkaipreviewengineerpython-agentic ·
crewAI 1.15.6: Bug fixes and documentation updates
Version 1.15.6 of crewAI addresses several bugs, including fixes for Anthropic tool-use detection and strict tool schema property names. It also improves error handling for crew and flow executions and agent loading. This patch primarily affects developers using crewAI for agent-based workflows, with no new features introduced.
patch - Pydantic AI Releases sdkaiengineerpython-agentic ·
Pydantic-AI v2.17.0: Cache OTel serialization, support arbitrary fields
This release of Pydantic-AI introduces support for arbitrary fields in `RequestUsage` and `RunUsage` to accommodate upcoming genai-prices. It also caches OpenTelemetry serialization per message, significantly reducing instrumentation costs from quadratic to linear. These changes benefit developers working with generative AI models, offering enhanced flexibility and improved performance.
feature patch - Pydantic AI Releases sdkaiengineerpython-agentic ·
pydantic-ai v2.16.0 adds Model Armor and new Gemini models
This release of pydantic-ai introduces several new features, including support for Google Cloud Model Armor and new Gemini models, alongside improvements to tool handling and error management. These updates benefit developers building AI applications with pydantic-ai, offering enhanced capabilities and robustness. Notably, it also includes dependency upgrades to address security alerts.
feature patch - Pydantic AI Releases sdkaideprecationengineerpython-agentic ·
pydantic-ai v2.15.0: Prompt caching, Mistral prompts, tool retries, and more
Pydantic-AI v2.15.0 introduces several new features including explicit prompt caching for OpenAI, inline text files in MistralModel prompts, and support for overriding tool-retry budgets. It also addresses multiple bug fixes related to hosted tool search, array handling, and durability capabilities. This release is primarily for developers and engineers using Pydantic-AI for LLM integrations and agent development.
feature patch - Google ADK (Python) Releases sdkaigcpdeprecationengineermediapython-agentic ·
google/adk-python v2.5.0: New features and numerous bug fixes
The google/adk-python library has released version 2.5.0, introducing significant new features such as enhanced Cloud Run integration, advanced agent capabilities, and improved workflow node validation. This release also addresses numerous bugs, including security improvements like mTLS support for API registries and GCP utilities. The update is relevant for developers using the ADK for building AI-powered applications, offering greater flexibility and robustness. Notably, it enables broader compatibility with a2a-sdk 1.x and enhances handling of various LLM interactions.
feature patch - Google ADK (Python) Releases sdkaiengineerpython-agentic ·
adk-python v1.36.2 patch: Gemini LLM connection logic update
This patch updates the custom Gemini LLM connection logic within adk-python to apply to all 3.x models, not exclusively 3.1. This change ensures broader compatibility for users leveraging different Gemini model versions. It is available as a patch release for users of the adk-python library.
patch - Pydantic AI Releases sdkaipreviewdeprecationengineerpython-agentic ·
pydantic-ai v2.14.1: Bug fixes for durable steps and wrapper-agent deprecations
This patch release for pydantic-ai includes two bug fixes addressing issues with fetching MCP instructions via durable steps and a temporary withdrawal of durable wrapper-agent deprecations. These changes aim to improve the reliability of the system by ensuring proper handling of durable operations. The fixes are available in version 2.14.1.
patch - Pydantic AI Releases sdkaipreviewdeprecationengineerpython-agentic ·
Pydantic-AI v2.14.0: Reasoning effort and durability improvements
This release introduces enhanced reasoning capabilities for Mistral models and new durability options, replacing deprecated agent wrappers. It also includes fixes such as the removal of specific shut-down Gemini models. These changes benefit developers integrating AI models who need more control over reasoning processes and durability strategies. The update is available now.
feature patch - Pydantic AI Releases sdkaiengineermediapython-agentic ·
pydantic-ai v2.13.0: New Instrumentation, Usage Metrics, and Bug Fixes
Pydantic-ai version 2.13.0 introduces new features for instrumentation and usage tracking, including the ability to omit model request parameters from spans and a new `cache_hit_ratio` property. It also enhances error handling for content filters and fixes several bugs related to streaming and temporal workflows. These updates are primarily relevant to developers integrating AI models via pydantic-ai, offering improved debugging and monitoring capabilities.
feature patch
About Python release tracking on ReleaseBytes
Python point releases, security fixes and end-of-life dates matter to almost every stack — including indirectly, through what Databricks runtimes and cloud services pin. ReleaseBytes tracks CPython releases and ecosystem news, and its EOL tracker shows exactly when each Python version stops receiving security fixes.
Frequently asked questions
How often are Python release notes updated on ReleaseBytes? ›
Continuously. ReleaseBytes monitors the official Python release channels around the clock and publishes a plain-English summary of each announcement shortly after it lands.
What kinds of Python changes does ReleaseBytes track? ›
New features, enhancements, bug fixes, security advisories, breaking changes, deprecations and end-of-life announcements. Every item is tagged by type so you can filter to just the changes that need action.
How can I get alerts for new Python releases? ›
Set up a free email or Slack alert filtered to Python, subscribe to the weekly digest, or follow the RSS feed. Teams can also install the ReleaseBytes GitHub App or connect via MCP.
Where does the Python release data come from? ›
From the official sources: Python language news, releases and ecosystem updates. Every item links back to the original vendor announcement.