About Opus 4.5, 4.6, 4.7, and 4.8
Opus 4.5 is Anthropic’s heavyweight frontier model, optimized for heavy-duty agentic workflows, complex software engineering, and deep research. It introduces a dynamic “Effort Control” parameter so you can minimize token spend on standard tasks or maximize reasoning depth for difficult problems — with state-of-the-art results at GPQA Diamond 88.9%. In DeepMask it has a 200K-token context window and excels at 3D visualization, financial modeling, and sustained autonomous coding sessions. Opus 4.6 (released February 4, 2026) takes the series further with “Adaptive Thinking” toggles across four effort levels (Low, Medium, High, Max), “Context Compaction” for near-infinite agent sessions, and native parallel sub-agent orchestration. Its GPQA Diamond score reaches 91.3%, and it introduces the ability to spin up independent sub-tasks in parallel — making it uniquely suited for cybersecurity investigations, complex codebase refactors, and high-stakes research where edge-case analysis is mandatory. Opus 4.7 (released April 15, 2026) refines the 4.6 architecture with persistent cross-session memory, a redesigned tool-use planner that cuts redundant calls, and a further drop in hallucination rate. Its GPQA Diamond score reaches 93.0%, context extends to 1M tokens in DeepMask, and it adds “Checkpoint Recovery” so long agent runs can resume from the last verified state after an interruption. Opus 4.8 (released June 10, 2026) is the current ceiling of the family, pairing “Continuous Self-Verification” — where the model audits its own intermediate steps in real time — with fully multimodal reasoning over text, images, and structured data. It reaches 94.5% on GPQA Diamond, handles up to 1M tokens of context in DeepMask, and extends parallel sub-agent orchestration with automatic load balancing across sub-tasks, making it the strongest option for mission-critical research, security, and large-scale engineering work.Key Capabilities
State-of-the-Art Software Engineering
Outperforms human candidates on elite engineering exams. Handles codebase migrations, complex refactoring, and 30-minute autonomous coding sessions.
Adaptive Thinking and Effort Control
Opus 4.5 through 4.8 offer configurable reasoning depth — minimize cost for routine tasks, maximize capability for demanding problems.
Parallel Sub-Agent Orchestration
Opus 4.6 and later natively spin up independent sub-tasks to run tools in parallel — essential for cybersecurity and large-scale coding investigations, with automatic load balancing added in 4.8.
Context Compaction
Opus 4.6 and later automatically summarize older context to sustain “infinite-feeling” agent sessions without losing the original goal-state.
Checkpoint Recovery
Opus 4.7 and 4.8 resume long agent runs from the last verified state after an interruption, backed by persistent cross-session memory.
Continuous Self-Verification
Opus 4.8 audits its own intermediate reasoning in real time and reasons natively over text, images, and structured data to catch errors before they propagate.
Best For
Choose Opus 4.5 for deep research, large-scale document analysis, enterprise automation, and production-grade coding agents — particularly where you need a balance of maximum capability and cost efficiency. Choose Opus 4.6 for the most complex tasks in your stack: multi-day codebase refactors, high-stakes legal or financial research, and cybersecurity defense workflows where parallel sub-agent execution makes a meaningful difference. Choose Opus 4.7 when you need persistent memory and resumable long-horizon agent runs with higher accuracy than 4.6. Choose Opus 4.8 for mission-critical work that demands the highest accuracy, real-time self-verification, and multimodal reasoning over text, images, and structured data. For tasks that don’t require Opus-level intelligence, Sonnet 4.5 or 4.6 offers strong performance at lower cost.Use Cases
- Deep research and analysis — Digest massive datasets, financial documents, and technical reports using up to 200K (4.5 and 4.6) or 1M (4.7 and 4.8) tokens of context.
- Production-grade coding agents — Build autonomous agents that create, test, and iterate on entire codebases with up to 75% fewer build/lint errors.
- Enterprise automation — Automate complex Excel workflows, financial modeling, and multi-agent systems that refine their own capabilities.
- Cybersecurity defense — Perform end-to-end vulnerability investigations with 90%+ success rates in blind tests (Opus 4.6).
- High-stakes research — Legal, financial, and scientific discovery where edge-case analysis and multi-step verification are mandatory.
Specifications
Try Opus in DeepMask →