Deprecation
Also written: Deprecate, Breaking Changes
Phasing out older AI models or features, signaling they will be removed and replaced by newer versions.
Deprecation means that a company is announcing that an older version of an AI model, software, or feature will no longer be supported or available in the future. This is a common practice in technology to manage resources and encourage users to adopt newer, often better, versions.
For example, a company might deprecate an older AI model that is less efficient or capable than its latest model. They will typically provide a timeline for when the older model will be completely shut down, giving users time to switch to the new one.
Deprecation is the formal process of marking a software component, API, model, or feature as obsolete, indicating that it will be removed from future releases or support cycles. This signals to users and developers that they should migrate to newer alternatives to avoid disruption. The process often involves a schedule for eventual retirement, allowing for a transition period.
In the context of frontier AI labs, deprecation commonly applies to AI models. Labs may deprecate older models to focus development and resources on newer, more powerful, or more cost-effective architectures. This also allows them to manage the complexity of maintaining multiple model versions and to implement new safety or performance standards that may not be compatible with older models.
Frontier AI labs are actively deprecating older models and features as they release new versions. This is necessary to manage complexity and focus resources on current technology. Users are being notified to update their integrations to avoid disruptions when these older components are retired.
- Upcoming deprecation of selected GitHub Copilot models in mid-NovemberMicrosoft · GitHub Changelog: Copilot ·
- Upcoming deprecation of selected GitHub Copilot models in mid-OctoberMicrosoft · GitHub Changelog: Copilot ·
- 1st Place Winner: Coding Agent Calls Developer to Resolve Code BlockAnthropic · DeepLearning.AI ·
- Introducing Graph API v26.0 and Marketing API v26.0Meta · Meta for Developers Blog ·
- An update on our model deprecation commitments for Claude Opus 3Anthropic · Anthropic Research ·
- Commitments on model deprecation and preservationAnthropic · Anthropic Research ·
- Function calling and other API updatesOpenAI · OpenAI News ·
- Anthropic Claude
- Anthropic is detailing its process for retiring Claude Opus 3, including commitments to preserving model weights and documenting preferences.
- Meta Meta AI
- Meta AI is updating its Graph and Marketing APIs to new versions, which involves changes for developers.
- Microsoft Copilot
- Microsoft Copilot is notifying users about the upcoming deprecation of selected GitHub Copilot models, urging them to update integrations.
- OpenAI ChatGPT
- OpenAI has made updates to its API, including changes that may affect how users interact with older functionalities.
- Retirement
- Retirement is the final act of shutting down a model or service, whe
- Obsoletion
- Obsoletion is the state of being outdated, while deprecation is the
- Versioning
- Versioning refers to the management of different iterations of a pro
The term 'deprecation' originates from software engineering, where it has long been used to signal that a feature or component is outdated and will be removed in a future release. It's a standard practice for managing software evolution and backward compatibility. Source ↗
Look for a "Deprecation Notices" or "API Updates" section on the website of an AI provider like OpenAI or Anthropic. Read about a specific model or feature that is being deprecated and note the timeline provided for its removal. Then, try to find documentation for the newer, recommended version to understand the differences.
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- PromptRole-play a student struggling with math for AI tutor simulationSimulate a student's math struggles and uncertainty for AI tutor evaluation. · 5 min
- TemplateReusable workflow templates for healthcare teamsTeams can create and share templates for common workflows like documentation or prior authorizations. · 15 min
- TemplateTemplate for building an AI tool tailored to your workA structured approach to defining and building a custom AI tool for specific business needs. · 15 min
- TemplateEvals CSV prompt list for agent skillsA CSV file template for listing prompts used in evaluating agent skills. · 20 min
- HeuristicDouble coding agent task completion with docs linksProviding coding agents with a documentation link significantly increases their task completion rate. · 2 min
Questions people ask
- Why do companies deprecate AI models?
- Companies deprecate AI models to focus resources on newer, more advanced versions, improve efficiency, and simplify maintenance. It helps them stay competitive and offer the best possible technology to users.
- How long does deprecation usually take?
- The timeline for deprecation varies greatly by company and product. It often involves a grace period, ranging from months to years, allowing users to transition to newer alternatives before the old version is permanently removed.
First seen on Subvolts Jun 13, 2023. Updated Oct 15, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.