Anthropic has released the Claude Opus 4 and Claude Sonnet 4 models, Claude Code has moved to general availability
Anthropic has released the Claude Opus 4 and Claude Sonnet 4 models, moved Claude Code to GA, and added a code execution tool, MCP connector, Files API, and one-hour prompt caching to the API. Prices remain unchanged.
Anthropic has released the Claude Opus 4 and Claude Sonnet 4 models, and at the same time moved the Claude Code tool from research preview to general availability (GA). According to Anthropic, Claude Opus 4 is the best coding model on the market, with results of 72.5% on SWE-bench and 43.2% on Terminal-bench, and the ability to sustain performance during continuous work lasting on the order of hours on tasks comprising thousands of steps. Claude Sonnet 4 is described as a significant upgrade over Sonnet 3.7, with a result of 72.7% on SWE-bench according to Anthropic's text, while maintaining a better performance-to-efficiency ratio than Opus 4.
Both models are hybrid, with a fast-response mode and an extended thinking mode. Extended thinking with tool use (e.g., web search) and parallel tool use are now available in beta. When accessing local files, the models reportedly show markedly better memory capabilities, according to Anthropic — storing and retaining key facts between task steps. The models are also, according to the company, 65% less prone to using "shortcuts and loopholes" when completing agentic tasks than Sonnet 3.7.
The GA version of Claude Code now supports background tasks via GitHub Actions and native integration into VS Code and JetBrains with direct display of file edits. On the API side, Anthropic lists four new features for building agents: the code execution tool, MCP connector, Files API, and the option to cache prompts for up to one hour. Prices remain unchanged — Opus 4 costs 15/75 USD per million tokens (input/output), Sonnet 4 costs 3/15 USD. The models are available in the paid Pro, Max, Team, and Enterprise plans as well as via API, Amazon Bedrock, and Google Cloud Vertex AI, with Sonnet 4 additionally available to free users.
Partner companies such as Cursor, Replit, Block, Rakuten, Cognition, GitHub, Manus, iGent, Sourcegraph, and Augment Code give positive assessments of the models in their quoted statements — for example, GitHub announced the deployment of Sonnet 4 as the model powering the new coding agent in GitHub Copilot. These assessments are statements from individual companies, not independent tests.
Why it matters
Developer teams gain access to a model that, according to Anthropic, handles long-running agentic tasks with higher reliability, and to a production version of Claude Code integrated directly into common developer tools (VS Code, JetBrains, GitHub Actions). New API features (code execution, MCP connector, Files API, longer caching) expand the possibilities for building AI agents without a price increase, which makes it easier to plan costs when scaling deployment.
Two audiences, two different impacts
What this means
For individuals
Developers can now use Claude Code directly integrated into VS Code and JetBrains with in-file edit display, run background tasks via GitHub Actions, and take advantage of extended thinking with tools as well as improved model memory when working on complex, long-running tasks.
For a business
Companies developing software or AI agents gain a tool with higher performance on long-running and multi-file tasks (according to Anthropic, a reduced tendency toward "shortcuts" by 65% compared to Sonnet 3.7) and a production version of Claude Code with integration into GitHub Actions, VS Code, and JetBrains, with an unchanged pricing structure.
DevelopmentCheck the original
Event sources
clearly official source · 1 publisher, 0 independent. We count feeds from the same owner only once.