changelogAnthropic

Anthropic Python SDK v0.97.0 Adds CMA Memory Feature in Public Beta

TL;DR

Anthropic has released version 0.97.0 of its Python SDK, introducing CMA Memory as a public beta feature. The update includes bug fixes for API spec errors, restored missing features, and performance improvements for multipart file requests.

1 min read
0

Anthropic Python SDK v0.97.0 Adds CMA Memory Feature in Public Beta

Anthropic released version 0.97.0 of its Python SDK on April 23, 2026, introducing CMA Memory as a public beta feature alongside bug fixes and performance improvements.

Key Changes

The primary addition in this release is CMA Memory, now available in public beta. The SDK update does not provide technical details about CMA Memory's functionality or implementation in the changelog.

The release addresses multiple API specification issues. Bug fixes include corrections to errors in the API spec and restoration of previously missing features that had been inadvertently removed.

Performance and Technical Updates

The SDK includes a performance optimization for handling multipart requests. According to the changelog, the client now optimizes file structure copying in multipart requests, though specific performance metrics are not disclosed.

Additional technical changes include:

  • Addition of a missing import statement
  • Enhanced bootstrap script robustness
  • Test dependency update to steady v0.22.1

What This Means

This SDK update appears to be a significant feature release for Anthropic developers, with CMA Memory being the headline addition. However, the changelog provides minimal context about what CMA Memory does or how it functions, leaving developers to reference separate documentation. The bug fixes addressing API spec errors and missing features suggest the previous version (0.96.0) may have had stability issues that are now resolved. The performance optimization for multipart requests could benefit developers working with file uploads or structured data submissions, though the actual performance impact remains unquantified.

Related Articles

benchmark

METR report: Anthropic's unreleased Model 4 scores 178.6 on AECI, ahead of Opus 5.5

METR's new report discloses three unreleased Anthropic models, Model 2, Model 3 and Model 4. Model 4 scores 178.6 on Anthropic's AECI index, about 3.9 points above Opus 5.5. Anthropic co-founder Sam McCandlish confirmed the report's main findings but said much information was removed.

analysis

METR Report: Anthropic's AI Development Pace May Have Doubled Due to Internal AI Use

A new report from AI research nonprofit METR says AI use within Anthropic may have doubled the company's pace of AI development. The rate at which Anthropic's models are improving as of October is roughly twice the rate between March 2024 and October 2025, according to The Information.

product update

Anthropic launches Personal Finance in Claude with read-only Plaid access for US Pro, Max and Ultra users

Anthropic announced Personal Finance in Claude, a feature that connects bank and credit card accounts through Plaid with read-only access. Conversations in the Personal Finance hub are not used to train Anthropic's models. It is available only to Pro, Max and Ultra subscribers in the US.

research

Anthropic Reports Bacterial and Phage Proteins Assemble Into Vault Particles in E. coli, With Claude Aiding a 3.2 Å Map

Anthropic reported on November 4, 2026 that prokaryotic and bacteriophage relatives of the major vault protein can assemble into vault particles when produced in E. coli. Claude helped the team map one bacteriophage-derived shell at a nominal resolution of 3.2 angstroms.

Comments

Loading...