MW Local AI Connector vs AI Provider for exo

This plugin registers exo as an AI provider in WordPress 7’s AI Client SDK and Connectors page. exo connects all your devices into an AI cluster, enabling you to run frontier models locally. It exposes an OpenAI-compatible API that this plugin connects to. Features: Registers exo as a WordPress AI provider OpenAI-compatible text generation via exo’s chat completions API Auto-detect active models from your running exo cluster Capability detection — displays model capabilities (Text, Code, Thinking, Vision) as badges “Connect & Detect” / “Save & Re-detect” connector flow Optional API key authentication with secure storage Configurable endpoint URL (default: http://localhost:52415) Settings integrated into WordPress 7’s Connectors page Choosing a Model: exo exposes every model in its catalog, but only models actively loaded on your cluster will respond. Use “Connect & Detect” on the Connectors page to discover which models are running. Recommended — Instruct models produce clean, usable output for WordPress AI features (title generation, content suggestions, etc.): Llama-3.2-3B-Instruct-8bit — ~3 GB, fast, great for short tasks Meta-Llama-3.1-8B-Instruct-4bit — ~4 GB, good balance of speed and quality Llama-3.3-70B-Instruct-4bit — ~35 GB, best quality, needs a larger cluster Avoid — Reasoning/thinking models (Qwen3.5, DeepSeek, GLM, Nemotron-Nano) spend most tokens on internal chain-of-thought, producing slow responses with minimal visible output. To load a model: exo run mlx-community/Llama-3.2-3B-Instruct-8bit Requirements: WordPress 7.0 or later PHP 8.3 or later A running exo cluster (see exo documentation) Configuration The plugin can be configured via the Connectors page or environment variables: AIPRFOEX_ENDPOINT — exo API endpoint (default: http://localhost:52415) AIPRFOEX_API_KEY — Optional API key for authentication AIPRFOEX_MODEL — Model name to use (auto-detected if empty) You can also define these as constants in wp-config.php.

MW Local AI Connector provides WordPress AI connectors for: Local AI: connect WordPress to a self-hosted OpenAI-compatible inference proxy on your own machine. Actual Computer: connect WordPress to the fixed Actual Computer API endpoint with your API key. The plugin integrates with the WordPress Connectors screen and provides a dedicated settings experience for each connector. See MW Local AI Connector in action: This plugin is an independent project by Matt Wiebe and is not affiliated with, endorsed by, or sponsored by Actual Computer, Tailscale, Ollama, LM Studio, or any other third-party service it can be configured to talk to. External services This plugin can connect to a user-controlled Local AI proxy and to the Actual Computer API. Connections only happen after you explicitly configure the matching connector in the WordPress admin and either save it or trigger an AI request through the WordPress AI Client. Local AI proxy (user-controlled) The Local AI connector sends requests to the OpenAI-compatible endpoint URL that you enter on its settings page. This endpoint is intended to point at a proxy you run on your own machine (for example, in front of Ollama, llama.cpp, LM Studio, vLLM, or multiple local providers at once). What it sends: the prompts, messages, provider-prefixed model selection, and any other AI parameters supplied by callers of the WordPress AI Client. If you configure an API key, it also sends that key as a Bearer token so the proxy can authenticate the request. It also calls /v1/models on this endpoint to populate the model selector in settings. When it sends: when an admin saves connection details (model list refresh) and whenever WordPress generates a response through this connector. Where it sends: the endpoint URL you configure on the Local AI settings screen. By default no endpoint is configured and no requests are made. Because the endpoint is one you supply, this plugin cannot link a generic terms of service or privacy policy on your behalf — the applicable terms are those of whatever software and network operator is running at that URL. If you expose the proxy through Tailscale Funnel, Tailscale’s terms and privacy policy are available at https://tailscale.com/terms and https://tailscale.com/privacy-policy. If you expose it through Cloudflare Tunnel, Cloudflare’s terms and privacy policy apply. Actual Computer API The Actual Computer connector sends requests to the fixed base URL https://api.actual.inc/v1, operated by Actual Computer. What it sends: the prompts, messages, model selection, and any other AI parameters supplied by callers of the WordPress AI Client, plus the bearer API key you enter on the settings page. It also calls /v1/models to populate the model selector. When it sends: when an admin saves connection details (model list refresh) and whenever WordPress generates a response through this connector. Where it sends: https://api.actual.inc/v1. Service provider: Actual Computer. Terms of service: https://actual.inc/terms. Privacy policy: https://actual.inc/privacy. If you do not configure either connector, this plugin makes no outbound network requests.