# Claude Opus 4.6 is INSANE! What's New? Source: https://www.youtube.com/watch?v=H0K7veXQnNA Recap page: https://rapidrecap.app/video/H0K7veXQnNA Generated: 2026-02-05T23:34:03.079+00:00 --- ## Quick Overview Claude Opus 4.6 significantly outperforms previous models and competitors like Gemini 3 Pro and GPT-5.2 across knowledge work (1606 Elo score), long-context retrieval (93.0% on 256k context), and software failure diagnosis (34.9% accuracy). New developer features include Adaptive Thinking, context compaction, 1M token context (beta), 128k output tokens, and Claude Code updates like Agent Teams and integrations with Excel and PowerPoint. **Key Points:** - Opus 4.6 achieved a knowledge work Elo score of 1606, surpassing Opus 4.5 (1416) and GPT-5.2 (1462). - In long-context retrieval (MRCR v2, 8-needle), Opus 4.6 achieved 93.0% accuracy with 256k context and 76.0% with 1M context. - The model excels at software failure diagnosis (OpenRCA) with 34.9% accuracy, significantly better than Opus 4.5 (26.9%). - New API features for developers include Adaptive Thinking for dynamic reasoning depth, four effort levels (low, medium, high, max), Context Compaction, and 1M token context (beta). - Opus 4.6 supports outputs up to 128k tokens and new capabilities in Claude Code, such as Agent Teams for coordinating multiple agents. - Anthropic also released research previews for Claude in Excel (handling long-running tasks) and Claude in PowerPoint (building/editing slides). - The pricing for Claude Opus 4.6 remains $5/$25 per million input/output tokens. ![Screenshot at 00:02: A bar chart comparing model performance on 'Knowledge work' using GDPVal-AA \(Elo scores\), showing Opus 4.6 leading with a score of 1606.](https://ss.rapidrecap.app/screens/H0K7veXQnNA/00-00-02.jpg) **Context:** This video announces the release of Claude Opus 4.6, Anthropic's latest and most capable model, detailing its performance improvements over previous Claude versions and competitors across several benchmarks. The presenter walks through key evaluation charts, highlights significant new features available via the API, and showcases new product integrations designed to enhance agentic and knowledge work capabilities. ## Detailed Analysis The video introduces Claude Opus 4.6, highlighting its superior performance across multiple benchmarks. In Knowledge Work (GDPVal-AA Elo scores), Opus 4.6 scored 1606, significantly beating Opus 4.5 (1416), Sonnet 4.5 (1277), Gemini 3 Pro (1195), and GPT-5.2 (1462). Agentic tasks also showed improvement; for example, Agentic coding (SWE-bench Verified) reached 80.8%, slightly behind Opus 4.5's 80.9% but ahead of others. In long-context retrieval (MRCR v2, 8-needle), Opus 4.6 achieved 93.0% accuracy at 256k context and 76.0% at 1M context, demonstrating strong resistance to context rot. Furthermore, Opus 4.6 achieved 34.9% accuracy in Software Failure Diagnosis (OpenRCA), outperforming Opus 4.5 (26.9%). New API features include Adaptive Thinking, which lets the model decide on deeper reasoning based on task complexity, and four Effort levels (low, medium, high/default, max) to control thinking depth, which can be dialed down from high to medium to prevent overthinking on simple tasks. Other updates include Context Compaction for summarizing older context, 1M token context (beta), and 128k output tokens. Claude Code now supports Agent Teams for coordinating parallel work. The model also shows substantial upgrades in office tools, with research previews for Claude in Excel and Claude in PowerPoint, capable of building slides and performing complex, long-running analyses automatically. ### Performance Benchmarks - Opus 4.6 leads Knowledge Work (1606 Elo) and excels in long-context retrieval (93.0% at 256k context) and software failure diagnosis (34.9% accuracy). ### API Feature Updates - Introduces Adaptive Thinking for dynamic reasoning depth, four Effort levels (low, medium, high, max), Context Compaction for older context summarization, 1M token context (beta), and 128k output tokens. ### Claude Code Enhancements - New research preview feature called Agent Teams allows coordinating multiple agents to work in parallel on complex tasks. ### Tool Integrations - Claude now works better with office tools, featuring research previews for Claude in Excel (handling long-running tasks) and Claude in PowerPoint (automating slide creation). ### Agentic Capabilities - Opus 4.6 surpasses previous models in various agentic benchmarks, including an 84.0% score on Agentic Search (BrowseComp) and 91.9% on Agentic Tool Use (Retail). ![Screenshot at 00:04: Bar chart comparing Opus 4.6 \(1606\) against competitors like GPT-5.2 \(1462\) in Knowledge Work Elo scores.](https://ss.rapidrecap.app/screens/H0K7veXQnNA/00-00-04.jpg) ![Screenshot at 00:23: Bar chart illustrating Opus 4.6's superior performance \(93.0% at 256k context\) in Long-context retrieval compared to Opus 4.5 \(76.0% at 1M context\).](https://ss.rapidrecap.app/screens/H0K7veXQnNA/00-00-23.jpg) ![Screenshot at 00:28: Software failure diagnosis \(OpenRCA\) chart showing Opus 4.6 achieving 34.9% accuracy, significantly higher than its predecessors.](https://ss.rapidrecap.app/screens/H0K7veXQnNA/00-00-28.jpg) ![Screenshot at 01:16: Demonstration of Claude in Excel, showing the AI analyzing workbook data and providing structured explanations and cell-level citations.](https://ss.rapidrecap.app/screens/H0K7veXQnNA/00-01-16.jpg) ![Screenshot at 03:33: Code snippet demonstrating the creation and assignment of two specialized agents \(Researcher and Writer\) to an AgentTeam.](https://ss.rapidrecap.app/screens/H0K7veXQnNA/00-03-33.jpg)