<p>Grant and Kyle dive into a comprehensive review and live test of the newly released Claude Opus 4.7, a cutting-edge large language model. This session explores its capabilities for coding and game dev, specifically referencing the &quot;Renaissance / Plan Final Fantasy Tactics RPG Game&quot; project. Discover how this ai model performs under pressure and its potential impact on game design workflows.</p><p><br></p><p>🔴 LIVE at 9:30AM PT / 12:30PM ET</p><p><br></p><p>Anthropic just dropped Claude Opus 4.7, and we’re putting it through the gauntlet in real time.</p><p><br></p><p>Join Grant Harvey (Lead Writer at The Neuron) for an unscripted, warts-and-all test of Anthropic’s newest flagship model.</p><p><br></p><p>What we’re testing</p><p>- Advanced coding on tasks Opus 4.6 struggled with</p><p>- New higher-resolution vision support for images up to ~3.75 megapixels</p><p>- File system-based memory across multi-session work</p><p>- The new xhigh effort level, which sits between high and max</p><p>- Claude Code’s new /ultrareview slash command</p><p>- Auto mode for longer, less-interrupted agent runs</p><p><br></p><p>Why this matters</p><p>Opus 4.7 is the first model Anthropic is releasing with its new automatic cyber safeguards, following last week’s Project Glasswing announcement.</p><p><br></p><p>It’s also the direct upgrade path from Opus 4.6 at the same price:</p><p>- $5 per million input tokens</p><p>- $25 per million output tokens</p><p><br></p><p>If you build on Claude, this is likely the model you’ll be using next.</p><p><br></p><p>What’s changing under the hood</p><p>- New tokenizer, where the same input can map to more tokens depending on content type, roughly 1.0x to 1.35x</p><p>- State-of-the-art score on GDPval-AA, a third-party evaluation of economically valuable knowledge work</p><p>- Better instruction following, which means prompts written for earlier models may now behave differently</p><p>- Improvements across finance agent evals, document reasoning, and long-context tasks</p><p><br></p><p>Bring your hardest prompts. We’ll run them live and show you what breaks, what shines, and whether it’s worth migrating today.</p><p><br></p><p>Watch part two, where Grant covers Codex for (almost) anything: https://youtube.com/live/OiRkwm3-og0</p><p><br></p><p>📰 Full writeup in tomorrow’s newsletter: </p><p>🐱 Subscribe to The Neuron (700K+ readers): https://www.theneuron.ai</p>