News from the AI & ML world

DeeperML

Michael Nuñez@venturebeat.com //

Anthropic's Claude 4 Excels in Coding Reasoning

Anthropic has recently launched its Claude 4 models, showcasing significant advancements in coding and reasoning capabilities. The release includes two key models: Opus 4, touted as the world's best model for coding, and Sonnet 4, an enhanced version of Sonnet 3.7. Alongside these models, Anthropic has made its coding agent, Claude Code, generally available, further streamlining the development process for users. These new offerings underscore Anthropic's growing influence in the AI landscape, demonstrating its commitment to pushing the boundaries of what AI can achieve.

Claude Opus 4 has been validated by major tech companies with Cursor calling it "state-of-the-art for coding," while Replit reported "dramatic advancements for complex changes across multiple files." Rakuten successfully tested a demanding 7-hour open-source refactor that ran independently with sustained performance. The models operate as hybrid systems, offering near-instant responses and extended thinking capabilities for deeper reasoning. Key features include enhanced memory, parallel tool execution, and reduced shortcut behavior, making them more reliable and efficient for complex tasks.

Additionally, Anthropic is adding a voice mode to its Claude mobile apps, allowing users to engage in spoken conversations with the AI. This new feature, currently available only in English, is powered by Claude Sonnet 4 and offers five different voices. Interestingly, Anthropic is leveraging Elevenlabs technology for speech features, indicating a reliance on external expertise in this area. Users can seamlessly switch between voice and text during conversations, and paid users can integrate the voice mode with Google Calendar and Gmail for added functionality.
Original img attribution: https://venturebeat.com/wp-content/uploads/2025/05/nuneybits_Vector_art_of_a_modern_computer_with_computer_code_on_50902dd1-2dee-4268-bf6f-2c0cd89cd43c.webp?w=1024?w=1200&strip=all
ImgSrc: venturebeat.com

Share: bluesky twitterx--v2 facebook--v1 threads


References :
  • bsky.app: #AI can now refactor code. Would you also have to use an AI to debug the refactored code? https://arstechnica.com/ai/2025/05/anthropic-calls-new-claude-4-worlds-best-ai-coding-model/ #ArtificialIntelligence
  • Last Week in AI: #210 - Claude 4, Google I/O 2025, OpenAI+io, Gemini Diffusion
  • AWS News Blog: Introducing Claude 4 in Amazon Bedrock, the most powerful models for coding from Anthropic
  • venturebeat.com: Anthropic's Claude Opus 4 codes seven hours nonstop, sets record SWE-Bench score and reshapes enterprise AI
  • Data Phoenix: Anthropic's newest Claude 4 models excel at coding and extended reasoning
  • thenewstack.io: Claude Opus 4 With Claude Code: A Developer Walkthrough
  • composio.dev: Comparison of Claude Code and OpenAI Codex.
  • Last Week in AI: Anthropic’s new Claude 4 AI models can reason over many steps, Google has 10000 announcements, OpenAI makes a big acquisition
  • www.zdnet.com: Anthropic's free Claude 4 Sonnet aced my coding tests - but its paid Opus model somehow didn't
Classification: