1. What is Dify? Who is it suitable for?
Dify is an open-source large language model application development platform, essentially an “AI low-code development tool.” Without writing complex code, you can build chatbots, knowledge base Q&A systems, or automated AI workflows.
Its core value is transforming complex AI model capabilities into a visual, easy-to-use interface.
Main Features Overview
- AI Chat Applications: Quickly build ChatGPT-like conversational systems
- RAG Knowledge Base: Accurate Q&A based on your own documents
- Workflow Orchestration: Logic branches, API calls, automation flows
- AI Agents: Tool-enabled intelligent agents for complex tasks
- API Integration: Easy integration with websites, bots, and business systems
👉 Simply put:
- Dify = AI application development platform
- OpenClaw = model management platform
- The two can be used together
2. Pre-deployment Preparation: Configuration & Server Recommendations
1. Basic Requirements
| Item | Minimum | Recommended |
|---|---|---|
| CPU | 2 cores | 4+ cores |
| RAM | 4GB | 8GB+ |
| Disk | 10GB | 50GB+ |
| OS | Ubuntu 22.04 | Ubuntu 22.04 LTS |
| Docker | 19+ | Latest version |
👉 Practical experience:
- 4GB = “barely usable”
- 8GB = “smooth experience”
2. Recommended BandwagonHost Plans
BandwagonHost official website:
https://bandwagonhost.com/

| Plan | RAM | CPU | Disk | Traffic/month | Bandwidth | Recommended Location | Price | Purchase |
|---|---|---|---|---|---|---|---|---|
| KVM (Cheapest) | 1GB | 2 cores | 20GB | 1TB | 1Gbps | DC2 AO / DC8 ZNET | $49.99/year | Buy |
| KVM | 2GB | 3 cores | 40GB | 2TB | 1Gbps | $52.99/6 months $99.99/year |
Buy | |
| CN2 GIA-E (Recommended) | 1GB | 2 cores | 20GB | 1TB | 2.5Gbps | US DC6 CN2 GIA-E US DC9 CN2 GIA Japan SoftBank JPOS_1 Netherlands EUNL_9 US San Jose CN2 GIA Canada CN2 GIA |
$49.99/quarter $169.99/year |
Buy |
| CN2 GIA-E (AI Deployment) | 2GB | 3 cores | 40GB | 2TB | 2.5Gbps | $89.99/quarter $299.99/year |
Buy | |
| SLA (Uptime Guarantee) | 1GB | Dedicated 2 cores | 20GB | 1TB | 2.5Gbps | US DC5 SLA 99.99% uptime guarantee Free IP change every 2 weeks |
$65.89/quarter $239.99/year |
Buy |
| SLA (Business Hosting) | 2GB | Dedicated 3 cores | 40GB | 2TB | 2.5Gbps | $116.99/quarter $399.99/year |
Buy | |
| HK (Premium Choice) | 2GB | 2 cores | 40GB | 0.5TB | 1Gbps | Hong Kong CN2 GIA Tokyo CN2 GIA Osaka CN2 GIA Singapore CN2 GIA |
$89.99/month $899.99/year |
Buy |
| HK (Business Hosting) | 4GB | 4 cores | 80GB | 1TB | 1Gbps | $155.99/month $1559.99/year |
Buy | |
| OSAKA (Cost-effective High-end) | 2GB | 2 cores | 40GB | 0.5TB | 1.5Gbps | Japan Osaka CN2 GIA | $49.99/month $499.99/year |
Buy |
| OSAKA | 4GB | 4 cores | 80GB | 1TB | 1.5Gbps | $86.99/month $869.99/year |
Buy |
Selection recommendations:
- Testing: 2GB + Swap
- Practical use: 4GB
- Production: 8GB+
3. Swap (Memory Extension) Must Be Configured
If you are using a 2GB plan, you must add Swap:
fallocate -l 4G /swapfile chmod 600 /swapfile mkswap /swapfile swapon /swapfile echo '/swapfile none swap sw 0 0' >> /etc/fstab
Verify:
free -h
3. Install Docker Environment
1. Install Docker
apt update && apt upgrade -y curl -fsSL https://get.docker.com | bash usermod -aG docker $USER
After re-login, verify:
docker --version docker compose version
2. Install Tools
apt install -y jq
4. 3-Step Quick Deployment of Dify
1. Pull Latest Version
git clone --branch "$(curl -s https://api.github.com/repos/langgenius/dify/releases/latest | jq -r .tag_name)" https://github.com/langgenius/dify.git
2. Configure Environment Variables
cd dify/docker cp .env.example .env
Generate key:
openssl rand -base64 42
Replace in .env:
SECRET_KEY=
3. Start Services
docker compose up -d
Check status:
docker compose ps
5. Initialize Dify Admin Panel
Access:
http://YOUR_IP/install
Set:
- Admin email
- Password
Then enter the dashboard.
6. Connect Mainstream AI Models (Core Step)
Dify does not include models by default; you must connect via API.
1. DeepSeek (Best value)
- Configure API Key
- Use: deepseek-chat / deepseek-reasoner
2. OpenAI
- Supports GPT-4 / GPT-4o
- Custom Base URL supported
3. Claude (Anthropic)
- Strong long-text capability
4. Local Models (Ollama)
- Fully private
- Requires ≥8GB RAM
7. Build Private Knowledge Base (RAG)
1. Create Knowledge Base
Knowledge → Create
2. Upload Files
- Word
- Markdown
- TXT
3. Indexing Methods
| Type | Features |
|---|---|
| High Quality | Uses Embedding, best performance |
| Economy Mode | No API calls, low cost |
4. Use in Applications
Create Chatbot → Add Knowledge → Set TopK
8. Security & Domain Optimization
1. Must-change settings
| Item | Recommendation |
|---|---|
| SECRET_KEY | Must change |
| Database password | Must change |
| Redis password | Must change |
2. Change Port
EXPOSE_NGINX_PORT=8080
3. Configure HTTPS
apt install nginx certbot python3-certbot-nginx
Request certificate:
certbot --nginx -d yourdomain.com
9. Upgrade Dify
git fetch origin git checkout latest version docker compose down docker compose up -d
⚠️ Always backup .env before upgrading
10. FAQ Troubleshooting
1. Startup failure
Cause: insufficient memory
Solution: add Swap or upgrade server
2. Cannot access
Check:
docker compose ps ss -tlnp ufw status
3. File upload failure
Modify:
UPLOAD_FILE_SIZE_LIMIT=50
4. Knowledge base not working
Cause: Embedding model not configured
11. Final Recommendations
Configuration Guide
| Scenario | Recommended Specs |
|---|---|
| Testing | 2 cores / 2GB + Swap |
| Daily use | 2 cores / 4GB |
| Production | 4 cores / 8GB+ |
Model Recommendations
- Best cost-performance: DeepSeek
- Best quality: GPT-4o / Claude
- Self-hosted: Ollama
12. FAQ
Q1: Does Dify require 4GB RAM?
Not necessarily, but below 4GB you must add Swap or it may crash.
Q2: Can it be deployed without Docker?
No. Official deployment only supports Docker.
Q3: Difference between Dify and n8n?
- Dify: AI application platform
- n8n: automation tool
Q4: Must I use BandwagonHost?
No, but CN2 routing performs better for users in China.
Q5: Is the knowledge base free?
Depends on whether you use Embedding models (e.g., OpenAI API).
This article corresponds to YouTube video:





