Opus 5 First Impressions: Anthropic Cooked Again!
Claude Opus 5 First Impressions: Benchmarks, Pricing, Agentic Coding & Real-World Demos In this quick first-impressions video, I test Anthropic’s new Claude Opus 5 and compare it to “Fable 5” across major third‑party benchmarks (including Cursor and Devin/Frontier Code), highlighting strong agentic coding results and improved performance at roughly half the price. I cover pricing (same as Opus 4.8), effort/thinking budget tradeoffs, and note Anthropic intentionally reduced cybersecurity capabilities. Then I run real demos on claude.ai and Claude Code (1M context): a Pokémon encyclopedia website, a crowd animation forming “Hello World, I’m Opus” with added camera controls, and an ISS tracker that takes a long time to generate but produces a detailed globe with tracking, footprint, and predicted path. I also point out a questionable benchmark chart detail and share thoughts on model tiers and a planned deeper follow‑up. LINK: https://www.anthropic.com/news/claude-opus-5 My voice to text App: whryte.com Website: https://engineerprompt.ai/ RAG Beyond Basics Course: https://prompt-s-site.thinkific.com/courses/rag Signup for Newsletter, localgpt: https://tally.so/r/3y9bb0 Let's Connect: 🦾 Discord: https://discord.com/invite/t4eYQRUcXB ☕ Buy me a Coffee: https://ko-fi.com/promptengineering |🔴 Patreon: https://www.patreon.com/PromptEngineering 💼Consulting: https://calendly.com/engineerprompt/consulting-call 📧 Business Contact: engineerprompt@gmail.com Become Member: http://tinyurl.com/y5h28s6h 💻 Pre-configured localGPT VM: https://bit.ly/localGPT (use Code: PromptEngineering for 50% off). Signup for Newsletter, localgpt: https://tally.so/r/3y9bb0 00:00 Opus 5 00:42 Benchmark Wins Explained 00:59 Agentic Coding Breakthrough 01:30 Third Party Coding Scores 03:18 Pricing Availability Safety 04:54 Hands On Creative Demos 08:06 ISS Tracker Verdict Wrap
