Skip to content

Choosing a Model

WP-AutoInsight supports multiple AI providers and dozens of models. This guide explains how to pick the right one for your content goals and budget.

Go to WP-AutoInsight > Connections > API Keys. Models are shown as cards organized by provider. Click any card to select it as your active model — the selection saves automatically.

Each model is labeled with a cost tier so you can set expectations before generating:

Fast, inexpensive models. Good for high-volume drafts, summaries, and content where speed matters more than depth.

Examples: GPT-4.1 Mini, Gemini Flash, Claude Haiku.

Typical cost: Fractions of a cent per post.

The balanced middle tier — capable enough for professional blog posts, SEO-focused content, and structured writing.

Examples: GPT-4.1, Gemini Pro, Claude Sonnet.

Typical cost: $0.01–$0.05 per post.

The highest-quality models for complex writing, deep reasoning, and polished long-form content.

Examples: GPT-4.1 Pro, Claude Opus.

Typical cost: $0.10–$0.50 per post.

GoalRecommended tier
Quickly filling out a content calendarEconomy
Regular blog posts for a business siteStandard
Flagship articles, pillar contentPremium
Bulk generation (dozens of posts at once)Economy
Research-heavy posts with citationsPerplexity (any tier)

Each model card shows an estimated cost per post based on a typical content length. This is a rough guide — actual costs depend on your token limit setting and how long your prompts and responses are.

To check your real spending, visit your provider’s usage dashboard directly. Links are in the Understanding Costs page.

Image models are separate from text models. You configure them in Media > Images and Media > Infographics:

  • DALL-E 3 (OpenAI): High-quality, creative images. Requires an OpenAI API key.
  • Gemini Image: Fast and cost-effective. Requires a Gemini API key.
  • Stability AI: Most flexibility for artistic styles. Requires a separate Stability AI key.

Your selected text model has no effect on image generation — they run independently.

You can change your active model at any time from Connections > API Keys. The new selection takes effect immediately for all subsequent jobs. Existing scheduled posts and queued jobs will use the model that was active when they were created.