Claude News 07/24/2026 AI Rating: High

Anthropic Unveils Claude Opus 5, a New Flagship Model for Coding and Knowledge Work

#Claude#Opus#Anthropic#Models

Anthropic has announced Claude Opus 5, the latest addition to its Opus line. Built to be thoughtful and proactive, it posts state-of-the-art results on coding and knowledge-work evaluations. Pricing stays unchanged from the previous Opus 4.8, and the model rolled out simultaneously across Claude.ai, Claude Code, Claude Cowork, and the API.

Details

  • Positioning: Strengthened verification and iterative problem-solving, plus better visual output quality. Designed with a thoughtful, proactive character
  • Benchmarks: Tops all competitors on Frontier-Bench v0.1, doubling Opus 4.8’s score at lower cost. On CursorBench 3.2 it comes within 0.5% of Claude Fable 5’s peak performance at half the cost. Scores three times higher than the next-best model on ARC-AGI 3, roughly 1.5x the completion rate of the next competitor on Zapier AutomationBench, and beats Fable 5’s OSWorld 2.0 result at about a third of the cost
  • Domain gains: Organic chemistry tasks improved 10.2 percentage points over Opus 4.8, and protein-analysis tasks improved 7.7 percentage points
  • Pricing unchanged: $5 per million input tokens, $25 per million output tokens β€” identical to Opus 4.8. A fast mode runs at 2.5x the speed for 2x the base price
  • Availability: Claude.ai, Claude Code, Claude Cowork, and the Claude API (model name claude-opus-5) all launched simultaneously
  • Safety: Anthropic reports improved alignment and safety metrics, along with lower variance in behavior on agentic tasks
  • Use cases: Examples cited include a trading firm building a market-data feed for a new exchange in a single session, a computer-vision pipeline that reconstructs a 3D CAD model from drawing images, and correctly identifying the root cause of an open-source bug that a competing model only patched at the surface level

How to try it

  • Select β€œClaude Opus 5” from the model picker in Claude.ai, Claude Code, or Claude Cowork
  • When using the API, specify the model name claude-opus-5 in your requests
  • Enable fast mode for quicker responses (2.5x the speed at 2x the price)
Source https://www.anthropic.com/news/claude-opus-5

Related Articles

More Claude articles β†’
Claude News 08/14/2026 Read AI Rating: Medium

How Claude's Text Watermarking Works

Anthropic will start watermarking text from future Claude models using SynthID-Text, a technique that biases word-choice randomness in a detectable but statistically invisible way, driven primarily by the EU AI Act's transparency requirements for AI-generated content.
Claude News 08/10/2026 Read AI Rating: Medium

Anthropic Makes Claude Sonnet 5's Introductory Pricing Permanent

Anthropic confirmed that Claude Sonnet 5's launch pricing of $2/$10 per million input/output tokens will not increase to $3/$15 on September 1 as originally scheduled β€” the introductory rate is now the standard, permanent price.
Claude News 08/07/2026 Read AI Rating: Medium

Anthropic Cuts Claude Fable 5's Biology False-Positive Refusals by 85%

Anthropic rewrote the safety classifier 'constitution' governing when Claude Fable 5 routes biology-related queries to the less capable Opus 5, cutting unnecessary fallbacks by roughly 85% while still blocking genuinely dangerous dual-use requests.
Claude News 08/05/2026 Read AI Rating: Medium

Claude Enterprise Adds "Inference Hooks" β€” Real-Time Allow/Deny Checks Before Every Prompt Reaches the Model

Anthropic has launched Inference hooks in beta for Claude Enterprise, letting organizations route every governed prompt through their own AI security server for an allow-or-deny verdict before inference runs, across claude.ai, Cowork, and Claude Code.