AI Connector
Connect your own AI provider account: pick a provider, save its key, choose a model and set request limits. Requests are billed by your provider to your account.
- The AI Provider card: seven provider cards, each with its status
- The Connection card: API Key with Save Key
- Sidebar: the "How to connect" guide for the selected provider, and Reset to Defaults
AI Provider
Pick the service that writes your product content. Each provider keeps its own key and status, and you pay the provider directly for usage.
| Provider card | Default model | Suggested models |
|---|---|---|
| OpenAI (selected by default) · GPT models | gpt-4.1-mini | gpt-4.1-mini, gpt-4.1, gpt-4o-mini, gpt-4o |
| Anthropic · Claude models | claude-haiku-4-5-20251001 | claude-haiku-4-5-20251001, claude-sonnet-5-5, claude-opus-5-5 |
| Google Gemini · Google Gemini models | gemini-2.5-flash | gemini-2.5-flash, gemini-2.5-flash-lite, gemini-2.5-pro |
| Groq · Fast open models | llama-3.3-70b-versatile | llama-3.3-70b-versatile, llama-3.1-8b-instant |
| OpenRouter · Many models, one key | openai/gpt-4.1-mini | openai/gpt-4.1-mini, anthropic/claude-haiku-4.5, google/gemini-2.5-flash, meta-llama/llama-3.3-70b-instruct |
| Mistral · Mistral models | mistral-small-latest | mistral-small-latest, mistral-medium-latest, mistral-large-latest |
| Custom endpoint · Any OpenAI-compatible endpoint | None: choose one | None: read from the endpoint's model list, or type the ID |
Each provider card shows its status. The sidebar guide, How to connect {provider}, lists the steps for the selected provider (account, billing, creating a key) with links such as Get an API key, Billing and Model list.
Connection
Settings for the provider selected above. Save Changes keeps the provider, endpoint and model; keys are saved with their own button.
- API Key
Paste it and click Save Key: "Key saved for OpenAI. Loading its models..." It's encrypted on your server and never shown again: the note under the field says "A key is saved (encrypted). Paste a new key to replace it." Remove Key deletes it. Each provider keeps its own key. A key defined in wp-config.php hides the field.
- Endpoint Base URL
Shown for Custom endpoint only: an HTTPS OpenAI-compatible base URL, for example
https://gateway.example.com/v1. Enter it before you save the key; the key is tied to that host, and changing the host later means entering the key again. - Model
Provider default (…), a model from the list, or Other model ID... to type one. With a saved key the list shows the chat models your key can use (for example "42 models available to your key."); otherwise "Suggested models." Refresh models reloads it. Then Save Changes.
- Status
Test Connection tests the provider and model shown above, with that provider's saved key: "Connected to Anthropic using claude-haiku-4-5-20251001." Status: Connected, Not connected, Not tested, Invalid credentials, Rate limited or Error, with the provider's message when there is one.
Requests & Limits
How requests to the provider behave, and caps on requests sent with your API key. Cached answers are free and don't count.
| Setting | Default | Notes |
|---|---|---|
| Request Timeout (seconds) | 45 | 5–120. Raise it for slow models |
| Maximum Response Tokens | 1500 | 128–8192. Lower reduces cost; too low may cut descriptions short. Raise it for reasoning models such as gpt-5 |
| Creativity | Balanced | Precise, Balanced, Creative |
| Requests per User per Hour | 60 | 0 = no limit |
| Requests per Day (whole site) | 500 | 0 = no limit |
| Response Cache (minutes) | 60 | Identical requests reuse the answer for free; 0 disables |
| Log Provider Errors | On | Error type and HTTP status only (never content or keys), in WooCommerce → Status → Logs |
- Limits reset on the UTC clock (at the top of each hour, and at midnight). Failed requests count; cached answers, tests and model lists don't. When a limit is reached: "The request limit set on the AI Connector tab has been reached. Try again later or raise the limit."
- Rate-limited or overloaded providers (HTTP 429 or 5xx) are retried once automatically.
- The AI Connection card on the General tab shows the provider, model and status at a glance.