Claude Opus 5 leads the Artificial Analysis Intelligence Index at 61 with a million-token context, setting the capability bar the price cuts are chasing
Anthropic's Claude Opus 5, released 24 July, tops the Artificial Analysis Intelligence Index at 61 and its Agentic Index at 55.3, with a 1M-token context window. It is the capability reference point of the cycle — the ceiling against which OpenAI's price cuts and DeepSeek's cheap-and-fast releases are measured.
The two numbers tell different halves of the story. An Intelligence Index of 61 says Opus 5 leads on raw capability; an Agentic Index of 55.3 says it leads on the tool-using, multi-step work that is turning into the highest-value enterprise use of these models. Anthropic is not merely topping a static benchmark — it is topping the one that tracks where the revenue is going.
The million-token context is the enabling detail behind the agentic lead. Long-horizon agent work — reading a whole codebase, holding a long task in view — is bounded by how much the model can attend to at once, and a 1M window is what lets Claude Code and similar surfaces keep a large problem in context rather than paging it in and out.
The strategic frame is that capability and price have split into separate contests. Anthropic holds the capability crown while OpenAI cuts Luna 80% and DeepSeek ships at $0.14; the market is being fought on two axes at once. A buyer now chooses a point on a capability-versus-cost curve rather than a single best model, and the curve is moving under them weekly.
LLM-Stats — AI updates today (August 2026) → · Fello AI — Best AI models in August 2026: ChatGPT, Claude, Gemini and Grok →