Anthropic releases Claude Opus 4.8 with dynamic agent workflows and a 1M-token context window

Anthropic has released Claude Opus 4.8, a new version of its flagship model that arrives at the same base pricing as its predecessor while adding a cheaper fast mode and tools aimed at coordinating large-scale agentic work, according to reporting from TechCrunch and VentureBeat. The model is available across Anthropic’s products and APIs, and Google Cloud’s Gemini Enterprise Agent Platform lists it as generally available with a release date of May 28, 2026.

The update’s most notable new feature is a dynamic workflows tool, which TechCrunch says is designed to coordinate swarms of subagents. VentureBeat reports that the model can also spawn hundreds of parallel subagents for codebase-scale tasks, making it more useful for long-running software development, debugging, and other complex workflows that require many steps. That same report says the new fast mode is significantly cheaper, with Anthropic touting a 3x lower cost for that tier.

Pricing for the main Opus model remains unchanged from the previous version, with VentureBeat saying Opus 4.8 is offered at $5 per million input tokens and $25 per million output tokens. OpenRouter’s model listing matches those rates and describes Opus 4.8 as Anthropic’s most capable generally available model in the Opus family, with support for text, image, and file inputs and a 1 million-token context window.

Anthropic is also emphasizing safety and reliability improvements. The TechCrunch summary says the company is promising a more honest model, while VentureBeat says early testers found Opus 4.8 more likely to flag uncertainty about its work. That matters because these systems are increasingly being used for tasks where accuracy, transparency, and sustained reasoning are critical, especially in coding and enterprise settings.

The release comes as Anthropic continues to expand the use of its Opus line for professional and agent-based work. Google Cloud’s documentation says Opus 4.8 is optimized for coding, agents, computer use, and enterprise workflows, underscoring the company’s push to position the model as a tool for developers and businesses rather than a general chat assistant alone.

What happens next will depend on how quickly customers adopt the new model and whether Anthropic’s cheaper fast mode and subagent features translate into measurable productivity gains. For now, the launch signals a continued race among AI companies to deliver models that are not only more capable, but also more affordable and easier to direct across long, complicated tasks.