AI Providers

WriteWP supports 6 AI providers: Google Gemini, OpenAI, Ollama, OpenRouter, MiniMax, and Ollama Cloud. Primary + fallback failover included.

AI Providers

6 AI providers.
One plugin.

WriteWP connects you to every major AI platform — cloud, local, or hybrid. Pick your model, set your fallback, never stop shipping content.

Supported Providers

Google Gemini

gemini-2.5-flash · gemini-2.5-pro · Gemma 3/4

Fast, affordable, great for bulk rewrites

OpenAI

gpt-4o-mini · gpt-4.1 · o1 · o3

Industry standard, versatile, great quality

OpenRouter

Any OpenRouter-available model

Hundreds of models through one API key

Ollama (Local)

llama3:8b · Any pulled model

Zero API costs. Maximum privacy.

Ollama Cloud

Cloud-hosted Ollama models

Ollama models without local hardware

MiniMax

MiniMax-M2.7

Cost-effective, strong multilingual support

01 — Provider

Google Gemini

Endpoint: Google AI API

Google’s Gemini family delivers exceptional speed and cost efficiency. The 2.5-flash model is purpose-built for high-throughput tasks — bulk rewrites, batch title generation, mass meta description creation — at a fraction of the cost of larger models. When you need deeper reasoning, gemini-2.5-pro steps up with stronger analytical capabilities.

  • Lowest cost-per-token among cloud providers
  • gemini-2.5-flash optimized for speed and throughput
  • gemini-2.5-pro for complex reasoning tasks
  • Gemma 3/4 open models for flexible deployment
  • Ideal for sites processing hundreds of posts

Available Models

gemini-2.5-flash
gemini-2.5-pro
gemma-3
gemma-4

02 — Provider

OpenAI

Endpoint: OpenAI API

The industry standard. OpenAI’s model lineup covers every use case — from the ultra-affordable gpt-4o-mini for simple rewrites, to GPT-4.1 for nuanced, high-quality content generation, to the reasoning-focused o1 and o3 models for complex editorial decisions. If you want proven reliability with broad ecosystem support, this is your default.

  • Most widely tested and documented models
  • gpt-4o-mini for cost-effective everyday tasks
  • GPT-4.1 for premium content quality
  • o1 / o3 reasoning models for complex decisions
  • Largest community and support ecosystem

Available Models

gpt-4o-mini
gpt-4.1
gpt-4.1-mini
gpt-4.1-nano
o1
o3
o3-mini
o4-mini

03 — Provider

Ollama (Local)

Endpoint: http://127.0.0.1:11434

Run AI directly on your machine. Zero API costs. Maximum privacy. Your content never leaves your server — no data sent to third parties, no usage tracking, no rate limits. Pull any model from the Ollama library and start generating. Perfect for privacy-conscious publishers, agencies under NDA constraints, or anyone who wants full control over their AI stack.

  • $0 API costs — runs entirely on your hardware
  • 100% local — no data leaves your machine
  • No rate limits or usage caps
  • Default model: llama3:8b (4.7GB download)
  • Compatible with the full Ollama model library
  • Ideal for GDPR-sensitive and NDA-governed content

Popular Local Models

llama3:8b
llama3:70b
mistral:7b
phi3:mini
gemma2:9b
codellama:7b
qwen2:7b

04 — Provider

OpenRouter

Endpoint: OpenRouter (OpenAI-compatible)

One API key. Hundreds of models. OpenRouter aggregates models from every major provider — OpenAI, Anthropic, Google, Meta, Mistral, and more — into a single OpenAI-compatible endpoint. Swap models without changing code. Test new models instantly. Mix and match based on task complexity. If you want maximum flexibility without managing multiple API accounts, OpenRouter is the answer.

  • Access 200+ models through a single API key
  • OpenAI-compatible — drop-in endpoint swap
  • Compare model quality and cost side by side
  • Provider-agnostic — no vendor lock-in
  • Instant access to newly released models

Why OpenRouter

One key, many models
OpenAI-compatible format
No vendor lock-in
Pay-per-use across providers
Instant model switching
New models available day-one

05 — Provider

Ollama Cloud

Endpoint: Via API key

All the Ollama models you love, none of the hardware requirements. Ollama Cloud gives you access to cloud-hosted Ollama models — the same open-weight models (Llama, Mistral, Qwen, etc.) running on managed infrastructure. No GPU needed on your end. No model downloads. No local resource constraints. Just connect your API key and start generating.

  • No local hardware requirements — runs in the cloud
  • Same open-weight models as local Ollama
  • No model downloads or storage needed
  • Simple API key authentication
  • Bridge between local and cloud Ollama setups

Local vs Cloud

Local: $0 cost, max privacy
Cloud: $0 hardware, max convenience
Same model library
Same API structure
Switch anytime
Best of both worlds

06 — Provider

MiniMax

Endpoint: api.minimax.io (international, OpenAI-compatible)

MiniMax brings strong multilingual capabilities at competitive prices. Built by one of China’s leading AI companies, the MiniMax-M2.7 model excels at content generation across Chinese, English, Japanese, Korean, and other languages. The OpenAI-compatible endpoint means zero integration friction — just plug in your API key and go. Ideal for multilingual WordPress sites targeting Asian markets.

  • Exceptional multilingual support (CJK+ languages)
  • Cost-effective pricing structure
  • OpenAI-compatible endpoint — easy integration
  • Strong performance on Chinese-language content
  • International endpoint with global availability

Key Details

Model: MiniMax-M2.7
Format: OpenAI-compatible
Endpoint: api.minimax.io
Languages: 15+ supported
Region: International
Pricing: Highly competitive

Resilience

Primary + Fallback.
Your pipeline never stops .

WriteWP supports a primary and fallback provider configuration. When your primary provider returns an error, times out, or hits a rate limit, WriteWP automatically retries the request with your fallback provider. No manual intervention. No broken workflows. Your content pipeline keeps running.

Step 01

Request is sent to your primary provider — Gemini, OpenAI, or any configured provider.

Step 02

If the primary fails (error, timeout, rate limit), WriteWP detects the failure instantly.

Step 03

The request is automatically retried with your configured fallback provider. Content delivered. Pipeline intact.

Quick Guide

Which provider should I choose?

It depends on what matters most to you. Here’s a quick reference:

Budget

Gemini / Ollama

Quality

GPT-4.1

Privacy

Ollama Local

Variety

OpenRouter

Multilingual

MiniMax

Get Started

Ready to connect your AI?

Install WriteWP, pick your provider, and start generating content in minutes.

Install WriteWP →