• My Feed
  • Home
  • What's Important
  • Media & Entertainment
Search

Stay Curious. Stay Wanture.

© 2025 Wanture. All rights reserved.

  • Terms of Use
  • Privacy Policy
banner
Tech/Software

Anthropic launches Claude Opus 4.5 for autonomous coding

25 November 2025

—

News

Jasmine Wu

Anthropic released Claude Opus 4.5, which the company calls its best model for coding, agents, and computer use, alongside new features including unlimited conversations, desktop tools for Chrome and Excel, and autonomous coding capabilities.

Why it matters: The release represents a shift in how AI assistants handle extended work sessions, with features designed to let developers and knowledge workers maintain context across longer timeframes while the system handles implementation details.

What's new: Conversations can now continue indefinitely. The system automatically compresses earlier context, eliminating previous length limits that required users to start new chat sessions.

The big picture: The Claude Developer Platform adds three capabilities designed to extend autonomous operation:

  • Tool search lets Claude find appropriate functions automatically
  • Programmatic tool calling gives developers precise control over which tools activate when
  • Context compaction compresses earlier conversation into smaller memory space

Combined with effort control features, these updates allow Claude to work longer with reduced human intervention.

Details: Claude Code now operates in parallel windows, enabling users to run multiple sessions simultaneously - one writing backend code, another testing APIs, a third updating documentation - all while the user focuses on other tasks.

Plan Mode receives an upgrade with Opus 4.5. The system asks clarifying questions upfront, then works autonomously toward the stated goal.

By the numbers: According to Anthropic, Claude Opus 4.5 ranks among top models on SWE-bench, a benchmark evaluating multi-file bug fixing and agentic tool use.

Performance varies across different benchmarks. HumanEval measures short function correctness, while LiveCodeBench evaluates interactive, timed coding. Model variants with extended reasoning modes typically outperform base versions. Competing models from other providers also achieve high scores on coding benchmarks.

What's next: Claude for Chrome expanded to all Max plan users. Max, Team, and Enterprise subscribers can now access Claude directly within Excel spreadsheets.

The bottom line: The release focuses on extending how long AI assistants can work autonomously and maintaining context across longer interactions - capabilities that could change workflows for developers and knowledge workers who currently manage multiple tools and context switches.

Anthropic's Claude Opus 4.5 is available now. Chrome integration requires Max plan. Excel access available for Max, Team, and Enterprise subscribers.

Topic

OpenAI Market Competition

Microsoft writes 30% of its code with AI

3 days ago

Microsoft writes 30% of its code with AI

Google's infrastructure advantage just ended OpenAI's dominance

23 November 2025

OpenAI's Altman warns staff of financial risks after Gemini 3 launch

21 November 2025

What is this about?

  • Artificial Intelligence/
  • Software Development/
  • Technology Innovation/
  • Developer Tools
Tech/Software

Anthropic launches Claude Opus 4.5 for autonomous coding

25 November 2025

—

News

Jasmine Wu

Anthropic released Claude Opus 4.5, which the company calls its best model for coding, agents, and computer use, alongside new features including unlimited conversations, desktop tools for Chrome and Excel, and autonomous coding capabilities.

Why it matters: The release represents a shift in how AI assistants handle extended work sessions, with features designed to let developers and knowledge workers maintain context across longer timeframes while the system handles implementation details.

What's new: Conversations can now continue indefinitely. The system automatically compresses earlier context, eliminating previous length limits that required users to start new chat sessions.

The big picture: The Claude Developer Platform adds three capabilities designed to extend autonomous operation:

  • Tool search lets Claude find appropriate functions automatically
  • Programmatic tool calling gives developers precise control over which tools activate when
  • Context compaction compresses earlier conversation into smaller memory space

Combined with effort control features, these updates allow Claude to work longer with reduced human intervention.

Details: Claude Code now operates in parallel windows, enabling users to run multiple sessions simultaneously - one writing backend code, another testing APIs, a third updating documentation - all while the user focuses on other tasks.

Plan Mode receives an upgrade with Opus 4.5. The system asks clarifying questions upfront, then works autonomously toward the stated goal.

By the numbers: According to Anthropic, Claude Opus 4.5 ranks among top models on SWE-bench, a benchmark evaluating multi-file bug fixing and agentic tool use.

Performance varies across different benchmarks. HumanEval measures short function correctness, while LiveCodeBench evaluates interactive, timed coding. Model variants with extended reasoning modes typically outperform base versions. Competing models from other providers also achieve high scores on coding benchmarks.

What's next: Claude for Chrome expanded to all Max plan users. Max, Team, and Enterprise subscribers can now access Claude directly within Excel spreadsheets.

The bottom line: The release focuses on extending how long AI assistants can work autonomously and maintaining context across longer interactions - capabilities that could change workflows for developers and knowledge workers who currently manage multiple tools and context switches.

Anthropic's Claude Opus 4.5 is available now. Chrome integration requires Max plan. Excel access available for Max, Team, and Enterprise subscribers.

banner

Topic

OpenAI Market Competition

Microsoft writes 30% of its code with AI

3 days ago

Microsoft writes 30% of its code with AI

Google's infrastructure advantage just ended OpenAI's dominance

23 November 2025

OpenAI's Altman warns staff of financial risks after Gemini 3 launch

21 November 2025

What is this about?

  • Artificial Intelligence/
  • Software Development/
  • Technology Innovation/
  • Developer Tools