Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis of charts and documents, complex office deliverables, and coordinating parallel subagents.
Success rate of requests sent by the Amux AI gateway to each provider. Client errors are excluded.
Tok/s
24 hours · 10/15–10/22
Average output speed of streaming requests, in tokens per second (TPS).
TTFT
24 hours · 10/15–10/22
Average time to first token (TTFT) of streaming requests, measured from request dispatch to the first token received.
Activity
24 hours · 10/15–10/22
Relative request activity over time, broken down by serving provider. Indexed to the peak within the selected range (= 100); reflects trends and provider share, not absolute volume.
Related models
More models from Anthropic
Anthropic: Claude Haiku 5.51M contextfrom $0.015 / M tokens4 providersAnthropic: Claude Haiku 5.5 is Anthropic’s most affordable, fastest and most powerful small form factor model yet, built for high-volume, cost-sensitive workloads. It handles summarization, reduction, database query, classification, and subagent encoding tasks with low latency, making it particularly useful in real-time customer support, browser usage, and other speed-critical applications.
Anthropic: Claude Sonnet 5.51M contextfrom $0.1 / M tokens5 providersAnthropic: Claude Sonnet 5.5 is Anthropic’s upgraded Sonnet-class model that delivers greater performance than Claude Sonnet 5 while running more than 30% faster and at a lower cost for many workloads. Designed for agent-based coding, tool usage, and everyday professional tasks, it's ideal for developers and teams building fast and powerful AI agents.
Anthropic: Claude Opus 5.51M contextfrom $0.2 / M tokens5 providersClaude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code review and bug finding, financial and scientific analysis, and reading dense charts, diagrams, and screenshots, and it is more careful than its predecessor about only stating figures and citing sources it can back up.
The model completes comparable tasks in fewer steps and with fewer tokens than Opus 5, and reports on its work in plainer language, with clear updates on what it did, what it found, and what it needs from the user. Thinking is always adaptive, so effort is the main lever for trading off depth, latency, and cost, and lower effort settings remain effective for latency-sensitive workloads.