Entity
Anthropic – Claude Opus 4.7 (April 2026)
Anthropic launched Claude Opus 4.7 on April 16, 2026, with a 64.3% score on the SWE-Bench Pro coding benchmark—roughly 10 points above its predecessor—and improved visual reasoning (SiliconAngle, April 16). The release continues Anthropic's rapid model iteration cadence and supports its enterprise and government market expansion. This page tracks capability milestones for the Opus model line.
Importance: 70%Confidence: 90%Mentions: 1Updated: May 6, 2026
## Overview
AnthropicPBC launched Claude Opus 4.7 on April 16, 2026, the latest addition to its Claude large language model family (SiliconAngle, April 16). The model delivers significant improvements in coding and visual reasoning relative to its predecessor.
## Key Capabilities
- **Coding benchmark:** Opus 4.7 scored 64.3% on the SWE-Bench Pro programming benchmark, approximately 10 percentage points higher than Opus 4.6 (SiliconAngle, April 16)
- **Visual reasoning:** The new model also reportedly improves on visual reasoning tasks, though specific benchmark scores were not disclosed in initial coverage (SiliconAngle, April 16)
## Competitive Context
The SWE-Bench Pro score positions Opus 4.7 as a competitive frontier model for software engineering tasks. The coding improvement trajectory aligns with Anthropic's Claude Code enterprise growth and market share gains tracked in existing wiki entries.
## Relation to Anthropic's Broader Strategy
- Anthropic has been aggressively expanding its enterprise AI footprint, with Claude Code seeing significant enterprise adoption (per existing wiki entry).
- The Mythos restricted cybersecurity model and government access negotiations (tracked separately) suggest Anthropic is pursuing both commercial and government channels simultaneously.
- The model launch follows a period of significant corporate activity including the CoreWeave cloud infrastructure deal and Amazon's $25B investment.
## Strategic Significance
- **For attorneys:** Capability improvements in coding AI raise ongoing questions about AI-generated code ownership, copyright, and liability—an emerging legal framework tracked in existing entries.
- **For entrepreneurs:** A 10-point coding benchmark jump in a single model generation signals rapid capability progression that may accelerate automation of software development workflows.
- **For investors:** Continued frontier capability improvements support Anthropic's valuation and competitive positioning against OpenAI and Google DeepMind.
## Status
Claude Opus 4.7 was opened for access on April 16, 2026 (SiliconAngle, April 16).