View on Substack
Claude Opus 4.8 gives you more control than raw model choice. You set an effort level (high, extra, max), you can drop into Fast mode, and inside Claude Code you can run Dynamic Workflows. The setting matters more than the model name. The table below maps 6 common work scenarios to the right setting, with one sentence on why.
| Scenario | Setting | Why |
|---|---|---|
| Quick drafting or casual chat | Lower effort or smaller model | Opus 4.8 at high effort is overkill and burns rate limits. |
| Messy reasoning or multi-step debugging | High or extra effort | The judgment upgrade over Opus 4.7 pays off here. |
| Difficult long-running async work | Extra (xhigh) effort | This is the level Anthropic recommends for hard agentic tasks. |
| Fast iteration where speed beats cost | Fast mode (/fast) | About 2.5x faster and roughly 3x cheaper than the old Fast mode. |
| Codebase-scale migration, bug hunt, or security audit | Dynamic Workflows | Surface-area problems where parallel subagents win. |
| Final judgment where the cost of being wrong is high | Regular mode, max effort | Speed is not the priority when a wrong answer is expensive. |
Source: Product with Attitude. Tested May 2026.
Originally published on Substack: Claude Opus 4.8: What Changed, and How I'll Test It.