anthropic claude model-release ai-tools developer-tools

Claude Opus 4.7 Is Here — Anthropic Retakes the AI Crown

Anthropic releases Claude Opus 4.7 with major coding, vision, and long-task improvements. It narrowly retakes the lead as the most powerful generally available LLM.

AI Tutorials · 16 April 2026 · 2 min read

Quick answer

Anthropic released Claude Opus 4.7 on April 16, 2026 — its most capable publicly available model yet. It scores 87.6% on SWE-bench Verified (up from 80.8%), supports high-resolution images up to 2,576 pixels, and is designed to stay focused over hours-long agentic tasks. Pricing stays at $5/$25 per million tokens.

product-launch

Claude Opus 4.7 Is Here — Anthropic Retakes the AI Crown

Anthropic released Claude Opus 4.7 today, its most capable publicly available AI model. The release narrowly retakes the lead in the race for the most powerful generally available large language model, edging ahead of OpenAI’s GPT-5.4 and Google’s Gemini 3.1 Pro on several key benchmarks.

What’s New

The biggest gains are in software engineering. On SWE-bench Verified, which measures how well AI models can solve real-world coding tasks, Opus 4.7 scores 87.6% — up from 80.8% on Opus 4.6. On CursorBench, which tests autonomous coding performance, it jumped from 58% to 70%. On a professional knowledge work benchmark, Opus 4.7 achieved an Elo score of 1753, comfortably ahead of GPT-5.4’s 1674.

Opus 4.7 is also the first Claude model with high-resolution image support. It can now process images at up to 2,576 pixels on the long edge — more than three times the previous limit. That matters for enterprise users who need to analyse scanned contracts, technical drawings, or detailed charts.

Anthropic says the model is also better at staying focused during long-running tasks. One of the most common frustrations with AI agents is that they lose coherence or stop working when they hit errors mid-task. Opus 4.7 is designed to recover from tool failures and continue executing, rather than halting.

What This Means for You

If you use Claude for coding, writing, or analysis, you should notice meaningful improvements — especially on complex, multi-step tasks. The model is available now on claude.ai, the API, Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Foundry. Pricing stays the same as Opus 4.6: $5 per million input tokens and $25 per million output tokens.

The release also comes just days after Anthropic revealed Claude Mythos, a far more powerful model that the company is restricting to select partners for defensive cybersecurity. Opus 4.7 is the strongest model Anthropic is willing to make available to everyone.

For context on the broader AI model landscape, see our coverage of GPT-5.4’s launch or head to our getting started page if you’re new to AI tools.

Frequently asked questions

What is Claude Opus 4.7?

Claude Opus 4.7 is Anthropic's latest and most capable generally available AI model. It delivers major improvements in software engineering, vision, and long-running agentic tasks compared to its predecessor, Opus 4.6.

How much does Claude Opus 4.7 cost?

API pricing remains the same as Opus 4.6: $5 per million input tokens and $25 per million output tokens. It's also available through Claude Pro and Team subscriptions.

Is Claude Opus 4.7 better than GPT-5.4?

On several benchmarks, yes. Opus 4.7 outperforms GPT-5.4 on the Knowledge Work benchmark (Elo 1753 vs 1674) and leads on SWE-bench Pro for agentic coding. On graduate-level reasoning (GPQA Diamond), they are nearly tied at around 94%.

Where can I use Claude Opus 4.7?

Opus 4.7 is available now on claude.ai, the Claude API, Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Foundry.

What is the difference between Claude Opus 4.7 and Claude Mythos?

Claude Mythos is a more powerful model that Anthropic has restricted to select partners for cybersecurity use only. Opus 4.7 is available to everyone and is Anthropic's strongest publicly accessible model.