Anthropic releases Claude Opus 4-7 with token budgets and user profiles in Python SDK v0.96.0
Anthropic released Python SDK v0.96.0 on April 16, 2026, adding support for Claude Opus 4-7, a new flagship model. The update introduces token budgets for cost management and user profiles for personalized interactions.
Claude Opus 4.7 — Quick Specs
Anthropic releases Claude Opus 4-7 with token budgets and user profiles in Python SDK v0.96.0
Anthropic released version 0.96.0 of its Python SDK on April 16, 2026, adding support for Claude Opus 4-7 and introducing two new API features: token budgets and user profiles.
New model: Claude Opus 4-7
The update adds Claude Opus 4-7 to the available models in the SDK. Anthropic has not yet disclosed pricing, context window size, or benchmark performance for this model. The naming follows Anthropic's established pattern, suggesting this is the seventh iteration of the Claude Opus 4 series.
Token budgets for cost control
The SDK now supports token budgets, allowing developers to set limits on token consumption per request or conversation. This feature addresses a common pain point for production deployments where cost control is critical. Implementation details and exact API parameters are documented in the full changelog.
User profiles for personalization
User profiles enable applications to provide context about individual users to Claude, potentially improving response relevance and personalization. The feature likely allows storing user preferences, interaction history, or other metadata that Claude can reference during conversations.
Technical details
The release is available through pip install with pip install anthropic==0.96.0. The update maintains backward compatibility with existing implementations using earlier SDK versions. Developers can view the complete changelog comparing v0.95.0 to v0.96.0 on GitHub.
The SDK is Anthropic's official Python library for accessing Claude models through their API, used by developers building applications with Claude integration.
What this means
This release signals Anthropic's continued expansion of the Claude 4 Opus line with a new model variant, though the lack of disclosed specifications makes it difficult to assess performance improvements. The addition of token budgets and user profiles indicates Anthropic is addressing enterprise needs around cost management and personalization—two factors critical for production AI deployments. The token budget feature in particular suggests Anthropic is responding to customer requests for more granular cost controls, a feature that competing providers like OpenAI have offered through usage limits and rate limiting configurations.
Related Articles
Anthropic Launches Model Hardware Standard to Let AI Agents Control Lab Robots and Machines
Anthropic has released a research preview of the Model Hardware Standard (MHS), a protocol that lets AI agents discover and control physical devices like robotic arms and liquid handlers through a single interface. Built with HHMI Janelia Research Campus, the spec has been tested by Genentech, Carnegie Mellon, and QuEra, with Anthropic claiming it cuts hardware integration time from weeks to hours.
Salesforce and Anthropic Launch 'Claudeforce' Plugin With 37 Sales Skills for Claude
Salesforce and Anthropic unveiled Claudeforce, a plugin with 37 pre-built sales skills that lets Claude access Salesforce data and take actions like composing emails and updating records. The move comes as Salesforce shares have fallen 22% in 2026 amid investor concerns that AI could replace SaaS platforms.
Anthropic Unifies Memory Across Claude Chat and Cowork
Anthropic has merged Claude's memory system so information learned in chat is now automatically available in Cowork, and vice versa. The update also lets users view, edit, or delete stored memories, with sensitive topics excluded by default.
Anthropic Unifies Memory Between Claude Cowork and Claude Chat
Anthropic now shares a single memory system between Claude Cowork and Claude chat, so tasks and conversations inform each other. Memory is on by default for Free, Pro, and Max plans, but sensitive topics like health or beliefs remain excluded unless manually enabled.
Comments
Loading...