Descriptions:
Anthropic’s opening keynote at Code with Claude Tokyo 2026, delivered on June 12, serves as the official launch event for Claude Mythos 5 and Claude Fable 5 — the company’s two most capable models to date. Head of engineering for the Claude platform Caitlyn Les opened the session by announcing the models had been released hours earlier, with Fable 5 described as state-of-the-art on nearly all tested benchmarks across software engineering, knowledge work, scientific research, and vision. The keynote reported that year-over-year API volume on the Claude platform is up nearly 17 times.
Angela Jen, head of product for the Claude platform, introduced several new Claude Managed Agents capabilities shipping alongside the model release: scheduled agent deployments (allowing agents to run on user-defined cadences) and environment variable vaults (enabling agents to make authenticated API calls without direct key access). The broader Claude Managed Agents product is positioned as the packaging layer for Anthropic’s agentic harnesses, context management tools, and production-grade infrastructure.
Customer case studies anchored the business narrative: Rakuten described moving from Claude Code for individual productivity to deploying managed agent teams across engineering, product, sales, and finance — cutting major release cycles from quarterly to bi-weekly. Canva showcased how Claude-powered tools inside its design platform allow non-technical users to build interactive mini-apps from natural language. Anthropic’s head of product for research Diane also previewed what’s coming next for the model roadmap, framing the accelerating capability curve as an opportunity for developers to identify and ship product experiences that were impossible even weeks prior.
📺 Source: Claude · Published June 12, 2026
🏷️ Format: Keynote Launch
