Factory CLI Unveils Dual AI Model Selection for Optimized Development Workflows

San Francisco, CA – The factory cli tool has introduced a significant enhancement, enabling users to select distinct artificial intelligence models for its "plan mode" and "execute mode." This new functionality aims to optimize development workflows by allowing specialized AI capabilities to be deployed at different stages of a project, as highlighted by technologist Ian Nuttall.

Nuttall announced the feature on social media, stating, > "factory cli now lets you choose a different model for plan mode and execute mode." He emphasized the seamless transition, noting, "when you switch modes, the model changes too." This development reflects a growing industry trend towards more granular control over AI integration in software development.

According to Nuttall's tweet, his current preferred setup leverages "gpt-5-high" for strategic planning and "gpt-5-codex" for code generation and execution. The "gpt-5-high" model is likely tailored for complex problem-solving, architectural design, and high-level strategy, while "gpt-5-codex" suggests advanced capabilities in generating, debugging, and optimizing code. This specialization allows developers to harness the strengths of different AI systems for specific tasks.

Factory cli is recognized as an AI-powered development tool, supporting custom model configurations through its "Bring Your Own Key" (BYOK) system. The platform's "Droids" maintain context across workflows, facilitating automated code review, self-healing builds, and large-scale scripting for CI/CD pipelines. This new dual-model capability extends its existing suite of intelligent automation features.

Industry experts suggest that such specialized AI model deployment can lead to increased efficiency, higher quality output, and reduced development cycles. By separating the planning and execution phases with purpose-built AI, factory cli aims to provide a more robust and adaptable environment for modern software engineering teams. This approach aligns with the broader movement towards AI orchestration and intelligent automation in the tech sector.

The ability to dynamically switch between powerful models like those in the GPT-5 series for different operational modes signifies a leap forward in developer tooling. It empowers users to fine-tune their AI assistance, ensuring that the most appropriate intelligence is applied to each phase of the software development lifecycle, from initial concept to final implementation.