Skip to content

LLM Models

Frontend gives you access to multiple frontier AI models, allowing you to select the best model for your specific task. Each model has unique strengths in coding, speed, visual design, and image capabilities.

You can select which LLM model to use from the options menu in the chat panel:

  1. Open the chat panel
  2. Click the options or settings icon
  3. Choose your preferred model from the available options
  4. The selected model will be used for subsequent prompts

Frontend provides access to cutting-edge frontier AI models, carefully selected for both quality and speed. The current LLM models available are considered frontier AI models and selected for both quality and speed.

Recommended for: The strongest coding model

Claude Sonnet 4.5 is the most capable model for complex coding tasks, offering superior code generation, advanced debugging, and sophisticated architectural decisions.

Recommended for: Speed with strong coding and image skills

Claude Haiku 4.5 offers an excellent balance of fast response times, strong coding capabilities, and native image understanding. Ideal for most development tasks where you need quick responses without sacrificing code quality.

Recommended for: Visual design

GPT-5 excels at visual and design-oriented tasks, with superior visual design capabilities and excellent UI/UX implementation.

GPT-5 Mini provides faster response times than GPT-5 with good general-purpose capabilities, suitable for routine development work.

Recommended for: Absolute speed

ZAI GLM 4.6 is optimized for maximum speed, offering the fastest response times for quick code generation and rapid prototyping.

Important limitation: This model does not have native image capabilities and cannot process images in prompts.

We will continue to update and include new AI models as they become available. The model selection menu will always show current available options.

You can change models at any time during development. Each model has access to your entire project context, so switching models won’t affect continuity.

Consider your specific needs when selecting a model:

  • For the strongest coding: Claude Sonnet 4.5
  • For speed with strong coding and image skills: Claude Haiku 4.5
  • For visual design: OpenAI GPT-5
  • For absolute speed: ZAI GLM 4.6 (note: no image capabilities)
  • For lighter tasks: GPT-5 Mini

Having access to multiple frontier AI models gives you the flexibility to choose the right tool for each task, optimizing for speed, quality, or specialized capabilities as needed.