A key new feature in the Opus 4.6 API is "Adaptive Thinking," which lets developers specify the level of effort the model applies to a task. Setting the effort to 'max' forces the model to think without constraints on depth, a powerful but resource-intensive option exclusive to the new version.

To build a useful multi-agent AI system, model the agents after your existing human team. Create specialized agents for distinct roles like 'approvals,' 'document drafting,' or 'administration' to replicate and automate a proven workflow, rather than designing a monolithic, abstract AI.

The new multi-agent architecture in Opus 4.6, while powerful, dramatically increases token consumption. Each agent runs its own process, multiplying token usage for a single prompt. This is a savvy business strategy, as the model's most advanced feature is also its most lucrative for Anthropic.

Codex lacks a built-in feature for parallel sub-agents like Claude Code. The workaround is to instruct the main Codex instance to write a script that launches multiple, separate terminal sessions of itself. Each session handles a sub-task in parallel, and the main instance aggregates the results.

Effective prompting requires adapting your language to the AI's core design. For Anthropic's agent-based Opus 4.6, the optimal prompt is to "create an agent team" with defined roles. For OpenAI's monolithic Codex 5.3, the equivalent prompt is to instruct it to "think deeply" about those same roles itself.

Separating AI agents into distinct roles (e.g., a technical expert and a customer-facing communicator) mirrors real-world team specializations. This allows for tailored configurations, like different 'temperature' settings for creativity versus accuracy, improving overall performance and preventing role confusion.

In a head-to-head test to build a Polymarket clone, Anthropic's Opus 4.6 produced a visually polished, feature-rich app. OpenAI's Codex 5.3 was faster but delivered a basic MVP that required multiple design revisions. The multi-agent "research first" approach of Opus resulted in a superior initial product.

Instead of relying on a single, all-purpose coding agent, the most effective workflow involves using different agents for their specific strengths. For example, using the 'Friday' agent for UI tasks, 'Charlie' for code reviews, and 'Claude Code' for research and backend logic.

Agent Skills only load a skill's full instructions after user confirmation. This multi-phase flow avoids bloating the context window with unused tools, saving on token costs and improving performance compared to a single large system prompt.