WP Manifestindependent plugin directory
manifest / ai / ollama-chatbot-for-wordpress

108 AI Master

Test your local ollama with actual wordpress (live or local) and build custom intent detection system

by Sanjay · github.com/sanjaynegi88/ollama-chatbot-for-wordpress

★ 0stars
0forks

Install

No release zip yet. The repository archive installs, but the folder name will carry the branch suffix and updates will not flow:

wp plugin install https://github.com/sanjaynegi88/ollama-chatbot-for-wordpress/archive/refs/heads/main.zip
Readme github.com/sanjaynegi88/ollama-chatbot-for-wordpress ↗

A powerful AI chatbot plugin for WordPress that integrates local LLMs using Ollama. Run AI chatbots directly on your own server without relying entirely on cloud-based AI providers.

This plugin enables WordPress websites to deploy intelligent AI assistants powered by Ollama models such as Llama, Mistral, DeepSeek, Gemma, Phi, and more.

Features

  • WordPress AI chatbot integration
  • Ollama local LLM support
  • Privacy-focused AI processing
  • Supports multiple AI models
  • Real-time chat responses
  • Lightweight REST API architecture
  • Conversation memory support
  • Custom chatbot UI
  • Shortcode support
  • Streaming response support
  • Multi-model switching
  • Developer-friendly hooks & filters
  • AI assistant embedding anywhere on WordPress

Supported Models

Compatible with any Ollama-supported model including:

  • Llama 3
  • Mistral
  • DeepSeek
  • Gemma
  • Phi
  • CodeLlama
  • TinyLlama
  • Custom Ollama models

Requirements

  • WordPress 6.0+
  • PHP 7.4+
  • Ollama installed on server/VPS
  • Linux server recommended
  • Minimum 8GB RAM recommended for larger models

Installation

  1. Upload the plugin ZIP file
  2. Activate the plugin
  3. Install Ollama on your server
  4. Configure Ollama endpoint in plugin settings
  5. Select AI model
  6. Add chatbot shortcode to pages/posts

Install Ollama

Linux Installation

curl -fsSL https://ollama.com/install.sh | sh

Verify Installation

ollama --version

Download AI Models

Example:

ollama pull llama3

Other examples:

ollama pull mistral
ollama pull deepseek-coder
ollama pull gemma

Start Ollama Server

ollama serve

Default Ollama API URL:

http://localhost:11434

Plugin Configuration

Navigate to:

Settings → Ollama Chatbot

Configure:

  • Ollama API URL
  • AI Model
  • Temperature
  • Max Tokens
  • System Prompt
  • Streaming Responses
  • Chat Memory

Shortcode Usage

Embed chatbot anywhere using shortcode:

[ollama_chatbot]

Custom example:

[ollama_chatbot model="llama3" theme="dark"]

REST API Example

Send chat request:

POST /wp-json/ollama-chat/v1/message

Example payload:

{
  "message": "Hello AI assistant",
  "model": "llama3"
}

AI Features

  • Context-aware conversations
  • Multi-turn memory
  • AI assistant behavior prompts
  • Custom personality prompts
  • Knowledge assistant workflows
  • Coding assistant support
  • AI automation workflows

Privacy Focused

Unlike cloud AI services, Ollama allows:

  • Local AI processing
  • Private conversations
  • No external API dependency
  • Full data ownership
  • Offline AI capabilities

Performance

Performance depends on:

  • Server CPU/GPU
  • Available RAM
  • Selected AI model size

Recommended:

Model Size Recommended RAM
3B 8GB
7B 16GB
13B 32GB

Advanced Features

  • Streaming AI responses
  • Multi-chat sessions
  • Session history
  • AI prompt templates
  • Custom AI instructions
  • Frontend chatbot widget
  • AJAX-powered conversations

Developer Hooks

Before AI Request

do_action('ollama_chat_before_request', $payload);

Modify AI Response

apply_filters('ollama_chat_response', $response);

After Chat Completion

do_action('ollama_chat_after_response', $response);

Use Cases

  • Customer support chatbot
  • AI knowledge base assistant
  • Internal business assistant
  • AI coding helper
  • Educational AI tutor
  • AI content assistant
  • AI automation workflows

Security

  • Sanitized user inputs
  • Protected REST endpoints
  • Nonce validation
  • Rate limiting support
  • WordPress coding standards compliant

Troubleshooting

Ollama Server Not Responding

Verify Ollama is running:

ollama serve

Model Not Found

Pull the required model:

ollama pull llama3

Connection Refused

Verify the API URL:

http://localhost:11434

Ensure firewall/server allows local connections.

Future Improvements

  • RAG support
  • Vector database integration
  • PDF/document chat
  • Voice chatbot support
  • Multi-agent AI workflows
  • AI image understanding
  • OpenAI compatibility mode
  • LangChain integration

License

Licensed under GPL v2 or later.

Author

Built for WordPress AI automation, local LLM hosting, and privacy-focused AI chatbot experiences.