Anthropic Drops Claude Opus 5: Half the Price, Insane Coding, Zero AGI Hype
While the rest of Silicon Valley burns billions chasing sci-fi fever dreams, Anthropic just quietly shipped an AI upgrade designed for people who actually have jobs to do.
The new flagship release from Anthropic, named Claude Opus 5, keeps its price locked at $5 per million input tokens and $25 per million output tokens while matching the performance of their top-tier Claude Fable 5 at half the running cost.
On benchmark tests like Frontier-Bench v0.1 and CursorBench 3.2, the model landed within a microscopic 0.5% gap of its more expensive sibling. The main focus here is software engineering and complex workflows, cutting down on pointless extra steps during long-chain tasks like a senior dev who just wants to finish early on Friday.
It hasn't won every single crown, however. Anthropic openly admitted that Claude Opus 5 still trails behind Mythos 5 in offensive cybersecurity and biology benchmarks, keeping strict guardrails that block binary-based scanning and exploit generation while allowing standard vulnerability detection.
Internal behavioral audits show higher adherence to corporate guidelines than Opus 4.8 or Sonnet 5. The model now serves as the primary engine for Claude Max and Claude Pro users, prioritizing daily practical utility over expensive party tricks.
Tech giants will keep throwing trillions at hypothetical digital gods, but delivering actual utility at fifty percent off is what actually decides who survives the AI bubble.
Source: Anthropic
Comments
This is where the magic happens: AI reads your discussion and rewrites the article based on the most interesting comments. Each strong comment adds points to the meter below. Once the meter is full, the article updates live — no page reload needed.