Anthropic updates Python SDK with new analytics, spend-limit, and RBAC methods
Version 2.4.0 of the anthropic-sdk-python adds generally available endpoints for analytics, spend limits, and role-based access control, alongside expanded error types and documentation updates.
1 source · cross-referenced
- Adds generally available methods for analytics, spend limits, and RBAC in the Anthropic Python SDK
- Introduces new error types for tool-use and compaction limits
- Updates documentation to clarify Slack channel IDs, Compliance API scope, and role permissions
- Drops unused project product field in spend-limit increase responses
Version 2.4.0 of the Anthropic Python SDK introduces generally available methods for analytics, spend limits, and role-based access control (RBAC). These additions provide developers with programmatic control over usage tracking, budget enforcement, and access policies, which are commonly required in enterprise deployments.
The release adds new error types, including tool_use_limit_reached_error and compaction_limit_reached_error, to help applications handle quota and resource constraints more gracefully. It also makes the spend_limit field nullable in the spend-limit increase approve response, aligning the API with real-world usage patterns where limits may not always be enforced.
Documentation updates clarify the scope of the Compliance API, the meaning of Slack channel IDs, and the permissions granted by the capability_access_all role. The SDK’s README now points to the official documentation site instead of embedding details in the README, reducing duplication and improving maintainability.
The changelog notes internal process improvements, such as repository-specific review rules and clearer worker comment wording, but these do not affect external behavior. Example code and model IDs have also been updated to reflect current supported models.
- Nov 11, 2026 · Hugging Face
Hugging Face proposes token pooling-aware training to improve compressed multi-vector retrieval
Trust79 - Nov 11, 2026 · TechCrunch — AI
Fractile raises $720M Series C to develop AI chips and compiler stack
Trust79 - Nov 11, 2026 · The Verge — AI
Google’s Tabletop AI experiment lets users build custom collaborative apps via waitlist
Trust79