Anthropic Claude Opus 4.7

Claude Opus 4.7 is an Anthropic frontier model released April 16, 2026 under the model ID claude-opus-4-7, with a 1M-token context window, 128K max output on the Messages API, and pricing held at $5 per million input tokens and $25 per million output tokens — the same rate Opus 4.6 carried. The model introduced a new t Anthropic framed the release around rigor rather than raw score, positioning Opus 4.7 for long-running tasks with precise instruction-following and self-verification. New developer-facing surfaces include the xhigh effort level (between high and max), vision at 2,576px on the long edge, and Task Budgets in public beta

Claude Opus 4.7 is a proprietary multimodal model from Anthropic released April 16, 2026, designed for agentic coding, long-horizon task execution, and enterprise knowledge work, with text and vision inputs and a context window of up to 1,000,000 tokens. It achieves 87.6% on SWE-bench Verified and 78.0% on OSWorld-Veri Vision capabilities now support high-resolution inputs up to 2,576px on the long edge (3.75 megapixels), more than three times the resolution of prior Claude models, enabling finer interpretation of dense diagrams, UI screenshots, and document layouts. The release adds self-verification on long-running tasks and a task

Anthropic made Claude Opus 4.7 generally available on April 16, 2026, with the API identifier claude-opus-4-7 and per-token pricing held unchanged at $5 per million input tokens and $25 per million output tokens. The model is available across the Claude API, Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Foundry A temporary 7.5× premium request multiplier was offered for GitHub Copilot through April 30, 2026, and Opus 4.7 shipped with automated safeguards that detect and block inputs indicating prohibited or high-risk cybersecurity uses. Anthropic also launched a Cyber Verification Program to provide vetted access for legitima

Claude Opus 4.7 posts the strongest coding numbers of any generally-available frontier model at launch, with 87.6% on SWE-Bench Verified (up from 80.8% on Opus 4.6), 64.3% on SWE-Bench Pro (up from 53.4%), and 70% on CursorBench versus Opus 4.6's 58%. The model ID is claude-opus-4-7, available on the Claude API, Amazon The release emphasizes agent reliability over raw capability, with Anthropic reporting the highest quality-per-tool-call ratio they have measured, lower looping rates, and better recovery from mid-run tool failures. Two new developer surfaces ship: xhigh effort level (between high and max, and the new default in Claude

A third-party TypingMind cost calculator lists DigitalOcean's hosted "anthropic-claude-opus-4.7" SKU with concrete developer-facing metadata: a 1,000,000-token context window, 128,000 max output tokens, knowledge cutoff 2026-01-31, release dated 2026-04-16, and pricing of $5.00 per 1M input tokens and $25.00 per 1M out For developers integrating the DigitalOcean-hosted Claude Opus 4.7 endpoint, the calculator surfaces the practical cost drivers they need to budget against: the large 1M context window makes long-document and agentic workloads viable, while the $25/M output rate and non-trivial $6.25/M cache-write cost materially affec

Anthropic's Claude Platform release notes (July 24, 2026) confirm that Claude Opus 4.7 (claude-opus-4-7) remains available at standard speed on the Claude API, Amazon Bedrock, Google Cloud, and Microsoft Foundry, and that DigitalOcean's Inference Engine is part of the broader Anthropic model distribution footprint via For DigitalOcean Gradient AI Platform customers consuming the claude-opus-4-7 slug, this upstream deprecation is operationally relevant: applications routed through the DigitalOcean Inference Router against this model will need to drop the fast-mode parameter or migrate upstream. The release notes also situate Opus 4.7