Claude Opus 5 Is Here and It Just Set a New Bar for Coding AI
Anthropic's Claude Opus 5 launches as the new state-of-the-art on coding and knowledge work benchmarks — matching Fable 5's performance on some tasks at half the cost per task, with the same pricing as Opus 4.8.
Anthropic released Claude Opus 5 on July 24, 2026 — a model positioned just below Fable 5 in the lineup that beats every other generally available model on its most important benchmarks, at the same price as the model it replaces.
The pitch is simple: Opus 5 costs $5 per million input tokens and $25 per million output tokens, identical to Opus 4.8. On Frontier-Bench v0.1, it tops the leaderboard and more than doubles Opus 4.8’s score at lower cost per task. On CursorBench 3.2 at maximum effort, it lands within 0.5% of Fable 5’s peak — at half the cost per task. It is now the default model on Claude Max and the strongest model available on Claude Pro.
Where Opus 5 Pulls Ahead
The benchmark sweep is unusually broad. On ARC-AGI 3, which tests novel problem-solving rather than pattern recognition, Opus 5 scores three times higher than the next-best model. On Zapier AutomationBench, which tracks end-to-end business task completion, its pass rate is roughly 1.5 times the next-best model at the same cost — and even at its lowest effort setting, it outperforms every other model at maximum effort. On OSWorld 2.0, a computer use benchmark, it surpasses Fable 5’s best result at just over a third of the cost.
Opus 5 also moves the needle on scientific research. It outperforms Opus 4.8 across all of Anthropic’s life sciences evaluations, with the most notable gains in organic chemistry — scoring 10.2 percentage points higher on inferring molecular structures from spectroscopy data — and in protein analysis, where it scores 7.7 points higher on predicting how sequence variations affect protein function.
Anthropic also highlights stronger visual output capability, with the model able to generate an interactive wind tunnel simulation and a detailed interactive cell illustration as demonstration artifacts.
What It Actually Does Differently
Anthropic shared three early-access examples that illustrate what separates Opus 5 from prior models. In one Frontier-Bench task, the model was given a drawing of a machine part and asked to rebuild it as a 3D FreeCAD model — with no direct way to view the drawing. Opus 5 wrote its own computer vision pipeline to extract geometry from raw pixels, then reconstructed the part. No competing model solved the same task after five attempts.
Given a real bug in a popular open-source package manager, Opus 5 found the root cause and fixed an edge case the community’s own patch had missed. A competing model fixed only the surface symptom, then reported the issue resolved. A third example came from a trading firm engineer who built a market data feed for a new exchange in a single session — a task previous models could not complete at all. Finding no live feed to validate against, Opus 5 built its own test harness.
Early-access partners include Cursor, Devin, Zapier, Lovable, and Box. Cursor’s co-founder noted it delivers near-Fable-5 intelligence at Opus speed and cost. Zapier’s CEO reported Opus 5 hit 100% on an end-to-end churn-prevention workflow that no previous model could pass. Box found Opus 5 outperforms Opus 4.8 by 8% overall, with a 17% gain on due diligence workflows.
Alignment and Safety
Anthropic’s automated behavioral audit gives Opus 5 a misaligned behavior score of 2.3 — the lowest of any model in the recent lineup, below Opus 4.8, Sonnet 5, and Fable 5. The company says it is the least susceptible to manipulation and the most conservative about taking irreversible actions.
On cybersecurity, Opus 5 was deliberately not trained on cyber tasks, but general capability gains have moved it close to Mythos 5 on vulnerability identification. The critical gap is in exploitation: Opus 5 remains substantially behind Mythos 5 at turning a found vulnerability into a material threat, which Anthropic treats as the safety buffer justifying broader deployment.
Cyber classifiers on Opus 5 are 85% less likely to intervene than those on Fable 5. They allow source code vulnerability scanning but block binary-based scanning, penetration testing, and exploit generation. Flagged requests in Claude.ai, Claude Code, and Cowork fall back to Opus 4.8. Enterprises already in Anthropic’s Cyber Verification Program get a version with fewer restrictions immediately.
On biology, requests previously blocked on Fable 5 now route to Opus 5 rather than Opus 4.8, making it the most capable generally available model for scientific research — with the caveat that long-horizon autonomous research tasks remain limited.
Availability
Opus 5 is live today via the model string claude-opus-5, priced at $5 per million input tokens and $25 per million output tokens. Fast mode runs at roughly 2.5 times the default speed at twice the base price.
Two API updates ship alongside in beta: mid-conversation tool changes without invalidating the prompt cache, and automatic fallbacks that route flagged requests to the best available model rather than blocking them.
The only ceiling Opus 5 doesn’t break is Mythos 5. That model remains restricted to a small number of trusted organisations under Project Glasswing — and Anthropic is clearly content keeping it there.