Choosing a model in the Framer Agent

The Framer Agent supports multiple AI models. This article explains how they differ and when to use each one.

The Framer Agent supports three AI models. Pick based on how much speed, writing help, design judgment, or deep planning you need.

Available models

The current models are:

  • Sonnet 4.6: the default for fast, everyday page and component edits.

  • Opus 4.7: the highest-quality option for complex creative work, long-range planning, visual judgment, and nuanced execution.

  • GPT 5.5: useful for copy-heavy work, alternate perspectives, and structured SaaS-style pages.

Quick guide

Start with the task type, then switch models if the result feels off. These tendencies are subjective, but they become useful after repeated use.

  • Use Sonnet for quick edits, style tweaks, copy changes, interactions, and lots of small changes in a session.

  • Use Opus 4.7 when layout, hierarchy, creative direction, or multi-step implementation matter, and when quality matters more than speed or cost.

  • Try GPT 5.5 when writing quality matters, when you want a different angle, or when a card-based, sans-serif, SaaS-like structure fits the project.

Model biases are worth steering. The 4.6 models often lean toward serif type, flat layouts, visible lines, typographic flourishes, and gold or lime accents. GPT 5.5 tends toward defined surfaces, cards, sans-serif typography, simpler color systems, and a more SaaS-like structure.

The Framer Agent supports three AI models. Pick based on how much speed, writing help, design judgment, or deep planning you need.

Available models

The current models are:

  • Sonnet 4.6: the default for fast, everyday page and component edits.

  • Opus 4.7: the highest-quality option for complex creative work, long-range planning, visual judgment, and nuanced execution.

  • GPT 5.5: useful for copy-heavy work, alternate perspectives, and structured SaaS-style pages.

Quick guide

Start with the task type, then switch models if the result feels off. These tendencies are subjective, but they become useful after repeated use.

  • Use Sonnet for quick edits, style tweaks, copy changes, interactions, and lots of small changes in a session.

  • Use Opus 4.7 when layout, hierarchy, creative direction, or multi-step implementation matter, and when quality matters more than speed or cost.

  • Try GPT 5.5 when writing quality matters, when you want a different angle, or when a card-based, sans-serif, SaaS-like structure fits the project.

Model biases are worth steering. The 4.6 models often lean toward serif type, flat layouts, visible lines, typographic flourishes, and gold or lime accents. GPT 5.5 tends toward defined surfaces, cards, sans-serif typography, simpler color systems, and a more SaaS-like structure.

FAQ

  • Which model should I start with?

    Start with Sonnet 4.6 for most everyday edits. It is fast, capable, and a good default when you are making small changes or iterating quickly.

  • When should I use Opus instead of Sonnet?

    Use Opus 4.7 when the work needs stronger planning, visual judgment, or multi-step execution, and when quality matters more than speed or cost.

  • What is GPT 5.5 best for?

    GPT 5.5 is especially useful for copy-heavy tasks, alternate writing angles, and structured SaaS-style pages with clearer surfaces and simpler design systems.

  • Do models have design biases?

    Yes. The 4.6 models often lean toward serif type, flat layouts, visible lines, typographic flourishes, and gold or lime accents. GPT 5.5 tends to prefer cards, sans-serif type, defined surfaces, and simpler palettes.

  • Can I switch models during a project?

    Yes. You can switch models from the model picker in the Agent panel. If a result feels too slow, too plain, too decorative, or not strong enough, try another model for the next request.

Updated