Skip to main content

​
Overview

Configure which AI models are available in your app and how they behave. All model configuration lives in chat.config.ts under config.models.

​
Configuration Options

models: {
  providerOrder: string[];        // Provider sort order in model selector
  disabledModels: AppModelId[];   // Models hidden from all users
  curatedDefaults: AppModelId[];  // Default models enabled for new users
  anonymousModels: AppModelId[];  // Models available to anonymous users
  defaults: {                     // Default model for each task type
    chat: AppModelId;
    title: AppModelId;
    pdf: AppModelId;
    artifact: AppModelId;
    artifactSuggestion: AppModelId;
    followupSuggestions: AppModelId;
    suggestions: AppModelId;
    polishText: AppModelId;
    formatSheet: AppModelId;
    analyzeSheet: AppModelId;
    codeEdits: AppModelId;
    chatImageCompatible: AppModelId;
    image: ImageModelId;
  };
}

​
Provider Order

Control the display order of providers in the model selector: 
providerOrder: ["openai", "google", "anthropic", "xai"],
Providers not in this list appear after those listed.

​
Disabling Models

Hide specific models from all users: 
disabledModels: ["morph/morph-v3-large", "morph/morph-v3-fast"],

​
Curated Defaults

Models enabled by default for new users. Users can enable additional models in settings. 
curatedDefaults: [
  "openai/gpt-5-nano",
  "openai/gpt-5-mini",
  "google/gemini-2.5-flash-lite",
  "anthropic/claude-sonnet-4.5",
],

​
Anonymous Models

Restrict which models anonymous (non-authenticated) users can access: 
anonymousModels: [
  "google/gemini-2.5-flash-lite",
  "openai/gpt-5-mini",
  "openai/gpt-5-nano",
  "anthropic/claude-haiku-4.5",
],

​
Task-Specific Defaults

Different tasks use different models by default. This lets you optimize for cost and quality per use case.
TaskDescription
chatMain conversation model
titleGenerates chat titles
pdfPDF document analysis
artifactCreates artifacts (documents, code, sheets)
artifactSuggestionSuggests artifact improvements
followupSuggestionsGenerates follow-up questions
suggestionsGeneral suggestions
polishTextText refinement
formatSheetSpreadsheet formatting
analyzeSheetSpreadsheet analysis
codeEditsCode modifications
chatImageCompatibleChat with image input support
imageImage generation (see Image Generation)

​
AI Providers

Available providers should be listed in config.services.aiProviders: 
aiProviders: [
  "OpenAI", "Anthropic", "xAI", "Google", "Meta",
  "Mistral", "Alibaba", "Amazon", "Cohere", "DeepSeek",
  "Perplexity", "Vercel", "Inception", "Moonshot", "Morph", "ZAI",
],
Model selection and routing is handled through the Vercel AI Gateway.

​
Image Models

Image models use a separate ImageModelId type. Set the default: 
defaults: {
  image: "openai/gpt-image-1.5",
}

​
Model Types

Two types of models can generate images:
TypeType DefinitionDescription
Image modelImageModelIdStandalone image models from AI Gateway
MultimodalMultimodalImageModelIdLanguage models with image-generation tag
The union type AnyImageModelId covers both.

​
Multimodal Image Models

Language models tagged with image-generation in models.generated.ts can generate images inline. The generateImage tool automatically detects these and uses generateText with image output modalities instead of the dedicated generateImage API.
lib/models/image-model-id.ts
export function isMultimodalImageModel(modelId: string): modelId is MultimodalImageModelId {
  return multimodalImageModelIds.has(modelId);
}
When a user’s selected chat model is multimodal, it’s preferred over the default image model—keeping generation within the same model context. The image model is used by the generateImage tool for both generation and edit modes. See Image Generation for details.