Three significant developments converged on 23 June, reshaping the competitive landscape for frontier AI models.
Anthropic's Claude Fable 5 transitioned from free inclusion in Pro, Max, Team, and Enterprise subscriptions to paid usage credits at $10 per million input tokens and $50 per million output tokens — double the price of Claude Opus 4.8. The free trial window ran from 9 June through 22 June, but the government-ordered shutdown from 12 to 18 June meant subscribers received only four to five days of actual free access. The pricing positions Fable 5 as a premium tier above Opus, reflecting its superior capabilities on complex coding and reasoning tasks.
OpenAI launched Daybreak, a cybersecurity platform combining GPT-5.5 with specialised security capabilities. The product directly targets Anthropic's Project Glasswing, which operates across more than 150 partner organisations including AWS, Apple, Google, and Microsoft. Daybreak represents OpenAI's most aggressive move into the cybersecurity vertical, a market that Anthropic has dominated since Mythos 5's launch.
Google's Gemini 3.5 Pro entered its general availability window, with analysts tracking an expected launch between 23 and 30 June. The model features a 2-million-token context window — double Gemini 3.5 Flash's 1 million and the largest in any production frontier model — along with a Deep Think reasoning mode gated to the $250 per month Ultra subscription tier.
For context engineers, the simultaneous moves underscore the intensity of competition at the frontier. Three major providers are repositioning their most capable models within the same week, each targeting different competitive advantages: Anthropic on coding performance, OpenAI on cybersecurity, and Google on context length.
