Dify

Documentation Index

Fetch the complete documentation index at: https://docs.gravitex.ai/llms.txt

Use this file to discover all available pages before exploring further.

​

1. Quick Integration

​

1.1 Get an API Key

​

1.2 Open Dify Model Providers

1. Log in to Dify, click your avatar (top-right) → Settings
2. From the left menu, choose Model Provider
3. Find the OpenAI-API-compatible plugin in the list and install it

​

1.3 Add a Model

​

1.4 Context Length

​

2. Core Features

​

2.1 Chat Assistant

1. Create app → choose Chat Assistant template
2. Configure the system prompt:
   You are the GravitexAI integration assistant. Your job:
   - Answer questions about wiring up our API
   - Recommend models that fit the user's scenario
   - Refer the user to the docs or BD when unsure
   Keep tone friendly, professional, and concise.
   
3. Pick gpt-5.5 or claude-opus-4-7 as the model
4. Suggested parameters: temperature = 0.7, max_tokens = 2000

​

2.2 Workflow Application

• Intent classification: gemini-3.5-flash (high throughput, low latency)
• KB retrieval / embedding: text-embedding-3-large or gemini-embedding-001
• Long-doc reasoning: claude-opus-4-7 (1M context, strong reasoning)
• Code generation: gpt-5.5 or qwen3-coder-plus

​

2.3 Knowledge Base Q&A (RAG)

1. Create a knowledge base → upload documents (PDF / Word / Markdown / TXT)
2. Embedding model: text-embedding-3-large recommended
3. Chunking: auto-split by paragraph, ~500 tokens per chunk
4. Reference the KB in the app
5. Retrieval parameters:
   • Top-K: 3–5
   • Similarity threshold: 0.7
   • Reranker: enabled (significantly improves relevance)

​

3. Application Templates

type: chat assistant
model: gpt-5.5
system_prompt: |
  You are a professional AI customer support agent. Responsibilities:
  - Answer user questions
  - Provide product information
  - Handle post-sales support
  Keep tone friendly and professional.
temperature: 0.7
max_tokens: 2000

type: workflow
input: uploaded document
pipeline:
  1. Document parsing (Dify built-in parser)
  2. Long-form summary (claude-opus-4-7, 1M context)
  3. Structured highlight extraction
  4. Generate report
output: Markdown report

type: chat assistant
model: claude-opus-4-7
system_prompt: |
  You are a professional coding assistant, specialized in:
  - Writing and refactoring code
  - Debugging
  - Architecture design
  - Best-practice recommendations
  Produce clear, runnable code with a brief explanation.
temperature: 0.3

​

4. Advanced Features

​

4.1 Call Dify Apps via API

import requests

url = "https://your-dify-instance/v1/chat-messages"
headers = {
    "Authorization": "Bearer YOUR_DIFY_APP_API_KEY",
    "Content-Type": "application/json",
}

payload = {
    "inputs": {},
    "query": "Introduce GravitexAI in one sentence.",
    "response_mode": "streaming",
    "user": "user_123",
}

resp = requests.post(url, headers=headers, json=payload, stream=True)
for line in resp.iter_lines():
    if line:
        print(line.decode("utf-8"))

​

4.2 Multimodal Input (Vision)

{
    "inputs": {
        "image": "data:image/jpeg;base64,...",
        "instruction": "Identify the key information in this image."
    },
    "query": "Describe the image in detail and provide actionable insights."
}

​

4.3 Batch Processing

1. Pick a fast/cheap model (gemini-3.5-flash, gpt 5.4 mini)
2. Cap concurrency in your Dify workflow to avoid hitting rate limits all at once
3. Enable result caching to avoid re-billing identical inputs

​

5. Model Selection

​

Cost optimization: dev vs prod

development:
  model: gemini-3.5-flash       # low cost, fast iteration
  max_tokens: 1000
  temperature: 0.7

production:
  model: claude-opus-4-7         # flagship intelligence, reliable
  max_tokens: 2000
  temperature: 0.3
  fallback: gpt-5.5              # GravitexAI auto-failover

​

6. Best Practices

​

6.1 Structured prompts

# Role
You are a professional [role here].

# Task
Help the user complete [the specific task].

# Output Format
1. Summary (≤ 100 words)
2. Detailed analysis (bullet points)
3. Recommended actions

# Constraints
- Be factual; flag uncertainty explicitly
- Keep total length under 500 words
- Respond in [language]

​

6.2 Workflow design

​

6.3 Observability

• ✅ User feedback (thumbs up/down)
• ⏱️ P95 latency
• 💰 Cost per call and daily/monthly usage curve
• ❌ Error rate and root cause distribution

​

6.4 Versioning

• Regularly export Dify app configs (JSON / YAML) as backups
• Roll out new versions via canary; keep at least N-1 for rollback
• Pin model versions in production to avoid surprise behavior changes

​

7. Troubleshooting

• Re-copy the API key from the GravitexAI Console
• Verify the account has balance
• Confirm baseURL is https://api.gravitex.ai/v1 (note the trailing /v1)

• Use the canonical model ID (e.g. gpt-5.5, not GPT-5.5)
• “Model Name” and “Model Name in API endpoint” must be identical

• Prefer Flash / Mini tier models
• Reduce max_tokens
• Turn on Dify response caching

​

Performance tuning template

caching:
  enabled: true
  ttl: 3600s
  key: identical input

concurrency:
  max_in_flight: 10
  queue_size: 100
  timeout: 30s

resources:
  memory: 2GB
  cpu: 80%

​

8. Deployment Recommendations

​

Production Docker Compose (self-hosted Dify)

version: '3.8'
services:
  dify-api:
    image: langgenius/dify-api:latest
    environment:
      - SECRET_KEY=your-secret-key
      - DB_HOST=postgres
      - REDIS_HOST=redis
      - OPENAI_API_KEY=sk-your-gravitex-key
      - OPENAI_API_BASE=https://api.gravitex.ai/v1
    depends_on:
      - postgres
      - redis

  dify-web:
    image: langgenius/dify-web:latest
    ports:
      - "3000:3000"
    depends_on:
      - dify-api

  postgres:
    image: postgres:14
    environment:
      - POSTGRES_DB=dify
      - POSTGRES_USER=dify
      - POSTGRES_PASSWORD=password

  redis:
    image: redis:alpine

​

Security

• Store the API key in env vars or a secrets manager — never hardcode it inside Dify configs
• Terminate HTTPS in front of Dify (Nginx / Caddy / Traefik)
• Enable SSO / 2FA for the Dify admin console
• Keep base images and dependencies up to date

​

Health checks

import requests, time

def monitor_dify():
    try:
        r = requests.get("http://dify-api:5001/health", timeout=5)
        if r.status_code == 200:
            print("Dify healthy")
        else:
            print(f"Unhealthy, status: {r.status_code}")
    except Exception as e:
        print(f"Probe failed: {e}")

while True:
    monitor_dify()
    time.sleep(60)

​

9. Reconciliation & Insights