Anthropic / Artificial Analysis
Major
AnthropicClaudePricingBenchmarksAnthropic ships Claude Opus 5.5 at 40% lower cost, tops Artificial Analysis index
September 22, 20262 min read
Anthropic released Claude Opus 5.5 on September 22 — the first model since its public slowdown pledge. It matches or beats Fable 5.1 on most tasks, scores 58 on the Artificial Analysis Intelligence Index (new high), and costs $4/$20 per million tokens with cache reads at $0.20.
Why it matters
Ten days after calling for the industry to pace capability releases, Anthropic shipped a stronger, cheaper model that took the top independent leaderboard spot. The gap between safety rhetoric and product cadence is now visible in public.
Anthropic launched Claude Opus 5.5 on September 22, the first model in its Claude 5.5 family and the first major release since CEO Dario Amodei publicly called for the industry to slow the pace of capability gains.
The company says Opus 5.5 performs at the level of Claude Fable 5.1 on most work while costing roughly 40% less to run than Opus 5 on typical workloads. List pricing is $4 per million input tokens and $20 per million output tokens (20% below Opus 5). Cache reads dropped to $0.20 per million tokens (60% cheaper). Artificial Analysis ranks it first on its Intelligence Index at 58 — several points above GPT-6 Astra and Fable 5.1.
Published scores include 66.4% on Terminal-Bench 4.0 and strong results on agentic coding and knowledge-work evaluations. The model is available via the Claude API, AWS, Google Cloud and Azure. Fast mode (up to 2.5× speed) is offered at double the standard rate.