As a pioneer in the field of artificial intelligence, Anthropic officially released the Claude4 series models on May 22, 2025, including Claude Opus4 and Claude Sonnet4, opening a new chapter for coding, complex reasoning, and AI agents. Claude Opus4 is acclaimed as the best coding model globally, capable of continuously and efficiently handling complex long-term tasks and agent workflows. Meanwhile, Claude Sonnet4 represents a major upgrade from Sonnet3.7, offering more precise responses to user instructions and demonstrating excellent coding and reasoning capabilities.
Claude4 Series: Comprehensive Performance Upgrade
The Claude4 series is hailed as Anthropic's most powerful model combination to date. According to official introductions, Claude Opus4 is currently the top coding model globally, capable of continuous operation for hours in complex, long-running tasks, showcasing exceptional autonomous coding abilities. For instance, when handling complex open-source projects, Claude Opus4 can autonomously code for nearly seven hours, significantly enhancing AI's practicality in software development.
At the same time, Claude Sonnet4 represents a significant upgrade from Sonnet3.7, not only excelling in coding capabilities but also achieving notable improvements in reasoning abilities. Compared to its predecessor, Sonnet4 offers higher control during task execution, allowing users to flexibly adjust the model's behavior to meet different scenario needs.
SWE-bench Verified Certification: Industry Benchmark
The Claude4 series performed excellently in the SWE-bench Verified test, which is specifically designed to evaluate a model's ability to solve real-world software problems.
Official data shows that both Claude Opus4 and Claude Sonnet4 reached industry-leading levels, particularly showcasing unparalleled stability and precision in handling complex software problems. This achievement not only solidified Anthropic's leadership position in the AI coding domain but also provided developers with more reliable tools, supporting applications from simple scripts to large-scale project development.
Pricing and Availability: Targeting a Broader User Base
Anthropic announced that the Claude4 series was made available to all paid plan users on the day of release, while Claude Sonnet4 is also accessible to free plan users, greatly reducing the experience threshold. This strategy reflects Anthropic's determination to popularize AI technology, enabling more developers, researchers, and ordinary users to experience the power of cutting-edge AI.
Market Response: Mixed Reactions
Discussions on the X platform show a热烈response to the release of the Claude4 series. Some users expressed doubt about whether Claude4 could surpass OpenAI's latest models (such as o3), but more people are looking forward to its performance in coding and reasoning fields. Previously, Anthropic's CEO Dario Amodei hinted at the arrival of a new model "Neptune" in an interview, and this release of Claude4 seems to confirm this speculation, sparking widespread attention and discussion.
From the perspective of AIbase, the release of the Claude4 series is not only a reflection of Anthropic's technical strength but also opens up new possibilities for AI applications in software development, scientific research, and daily task processing. Its powerful coding capabilities and flexible control mechanisms may push AI from being an auxiliary tool to becoming a core productivity tool. However, facing continuous pressure from competitors like OpenAI, whether the Claude4 series can fulfill its promises in actual applications still requires time to verify.
Official blog: https://www.anthropic.com/news/claude-4