DeepSeek 深度求索

English

简体中文

English

简体中文

Appearance

Support & Help ​

Get the help you need with comprehensive support resources, documentation, and direct assistance from our expert team.

Overview ​

DeepSeek provides multiple support channels to ensure you get the help you need:

  • Self-Service Resources: Comprehensive documentation and guides
  • Community Support: Active community forums and discussions
  • Technical Support: Direct assistance from our technical team
  • Enterprise Support: Dedicated support for enterprise customers
  • 24/7 Availability: Round-the-clock support for critical issues

Support Channels ​

Documentation & Guides ​

Getting Started ​

Advanced Topics ​

Community Support ​

Community Forum ​

  • URL: https://community.deepseek.com
  • Response Time: Community-driven (typically within hours)
  • Best For: General questions, discussions, sharing experiences

Discord Community ​

  • URL: https://discord.gg/deepseek
  • Active Hours: 24/7 with peak activity during business hours
  • Channels:
    • #general: General discussions
    • #api-help: API-related questions
    • #showcase: Share your projects
    • #announcements: Official updates

GitHub Discussions ​

Technical Support ​

Support Ticket System ​

  • URL: https://support.deepseek.com
  • Response Time:
    • Free Tier: 48-72 hours
    • Paid Plans: 24 hours
    • Enterprise: 4 hours
  • Best For: Technical issues, billing questions, account problems

Email Support ​

Live Chat ​

  • Availability: Monday-Friday, 9 AM - 6 PM PST
  • Access: Available in the console
  • Best For: Quick questions and immediate assistance

Support Tiers ​

Free Tier Support ​

Included Support ​

  • Community Forums: Full access to community resources
  • Documentation: Complete access to all documentation
  • Email Support: Basic email support with 48-72 hour response
  • Knowledge Base: Searchable knowledge base
  • Status Updates: Service status and incident notifications

Limitations ​

  • No Phone Support: Email and community only
  • Standard Priority: Lower priority than paid plans
  • Business Hours: Support during business hours only

Enhanced Support ​

  • Priority Email: 24-hour response time guarantee
  • Live Chat: Access to live chat during business hours
  • Phone Support: Scheduled phone consultations
  • Technical Guidance: Advanced technical assistance
  • Integration Help: Assistance with integrations

Response Times ​

json
{
  "response_times": {
    "critical": "4 hours",
    "high": "8 hours",
    "medium": "24 hours",
    "low": "48 hours"
  },
  "availability": {
    "email": "24/7",
    "chat": "business_hours",
    "phone": "by_appointment"
  }
}

Enterprise Support ​

Premium Support Features ​

  • Dedicated Account Manager: Personal point of contact
  • 24/7 Priority Support: Round-the-clock critical issue support
  • Phone Support: Direct phone access to technical team
  • Custom SLA: Tailored service level agreements
  • Technical Reviews: Regular technical architecture reviews

Enterprise Support Channels ​

  • Dedicated Slack Channel: Direct access to support team
  • Emergency Hotline: 24/7 emergency support line
  • Technical Account Manager: Dedicated technical resource
  • Quarterly Business Reviews: Regular check-ins and planning

SLA Guarantees ​

json
{
  "enterprise_sla": {
    "critical_issues": {
      "response_time": "1 hour",
      "resolution_target": "4 hours",
      "availability": "24/7"
    },
    "high_priority": {
      "response_time": "2 hours",
      "resolution_target": "8 hours",
      "availability": "24/7"
    },
    "standard_issues": {
      "response_time": "4 hours",
      "resolution_target": "24 hours",
      "availability": "business_hours"
    }
  }
}

Common Issues & Solutions ​

Authentication Issues ​

Invalid API Key ​

bash
# Error: 401 Unauthorized
{
  "error": {
    "type": "authentication_error",
    "message": "Invalid API key"
  }
}

Solutions:

  1. Verify API key is correct and active
  2. Check for extra spaces or characters
  3. Ensure API key has required permissions
  4. Regenerate API key if necessary
python
# Verify API key
from deepseek import DeepSeek

try:
    client = DeepSeek(api_key="your-api-key")
    # Test with a simple request
    models = client.models.list()
    print("API key is valid")
except Exception as e:
    print(f"API key error: {e}")

Permission Denied ​

bash
# Error: 403 Forbidden
{
  "error": {
    "type": "permission_error",
    "message": "Insufficient permissions for this operation"
  }
}

Solutions:

  1. Check API key permissions in console
  2. Verify account has access to requested model
  3. Contact support to upgrade permissions

Rate Limiting ​

Rate Limit Exceeded ​

bash
# Error: 429 Too Many Requests
{
  "error": {
    "type": "rate_limit_error",
    "message": "Rate limit exceeded",
    "retry_after": 60
  }
}

Solutions:

  1. Implement exponential backoff
  2. Reduce request frequency
  3. Upgrade to higher rate limits
  4. Use batch processing
python
import time
import random

def make_request_with_backoff(client, request_func, max_retries=3):
    for attempt in range(max_retries):
        try:
            return request_func()
        except RateLimitError as e:
            if attempt == max_retries - 1:
                raise
            
            # Exponential backoff with jitter
            wait_time = (2 ** attempt) + random.uniform(0, 1)
            time.sleep(wait_time)

API Errors ​

Model Not Found ​

bash
# Error: 404 Not Found
{
  "error": {
    "type": "model_error",
    "message": "Model 'invalid-model' not found"
  }
}

Solutions:

  1. Check available models with client.models.list()
  2. Verify model name spelling
  3. Ensure account has access to model

Request Timeout ​

bash
# Error: 408 Request Timeout
{
  "error": {
    "type": "timeout_error",
    "message": "Request timed out"
  }
}

Solutions:

  1. Reduce request complexity
  2. Increase timeout settings
  3. Split large requests into smaller ones
  4. Check network connectivity
python
# Configure timeout
client = DeepSeek(
    api_key="your-api-key",
    timeout=60  # 60 seconds timeout
)

Integration Issues ​

SDK Installation Problems ​

bash
# Common installation errors
pip install deepseek-ai
# Error: Could not find a version that satisfies the requirement

npm install deepseek-ai
# Error: Package not found

Solutions:

  1. Update package manager: pip install --upgrade pip
  2. Check Python/Node.js version compatibility
  3. Use virtual environments
  4. Clear package cache

Import Errors ​

python
# Error: ModuleNotFoundError: No module named 'deepseek'
from deepseek import DeepSeek

Solutions:

  1. Verify installation: pip list | grep deepseek
  2. Check virtual environment activation
  3. Reinstall package: pip uninstall deepseek-ai && pip install deepseek-ai

Troubleshooting Guide ​

Diagnostic Steps ​

1. Check System Status ​

python
import requests

# Check DeepSeek status
status_response = requests.get('https://status.deepseek.com/api/v2/status.json')
status = status_response.json()

if status['status']['indicator'] != 'none':
    print(f"Service status: {status['status']['description']}")
else:
    print("All systems operational")

2. Verify Configuration ​

python
import os
from deepseek import DeepSeek

# Check environment variables
api_key = os.getenv('DEEPSEEK_API_KEY')
if not api_key:
    print("❌ DEEPSEEK_API_KEY not set")
else:
    print("✅ API key found")

# Test basic connectivity
try:
    client = DeepSeek(api_key=api_key)
    models = client.models.list()
    print(f"✅ Connected successfully. Available models: {len(models.data)}")
except Exception as e:
    print(f"❌ Connection failed: {e}")

3. Test API Functionality ​

python
def run_diagnostics():
    client = DeepSeek(api_key=os.getenv('DEEPSEEK_API_KEY'))
    
    tests = [
        ("List Models", lambda: client.models.list()),
        ("Simple Chat", lambda: client.chat.completions.create(
            model="deepseek-chat",
            messages=[{"role": "user", "content": "Hello"}],
            max_tokens=10
        )),
        ("Streaming", lambda: list(client.chat.completions.create(
            model="deepseek-chat",
            messages=[{"role": "user", "content": "Count to 3"}],
            stream=True,
            max_tokens=20
        )))
    ]
    
    for test_name, test_func in tests:
        try:
            result = test_func()
            print(f"✅ {test_name}: Passed")
        except Exception as e:
            print(f"❌ {test_name}: Failed - {e}")

run_diagnostics()

Performance Optimization ​

Reduce Latency ​

python
# Optimize for speed
response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[{"role": "user", "content": "Quick response needed"}],
    max_tokens=50,        # Limit response length
    temperature=0,        # Reduce randomness
    stream=True          # Stream for perceived speed
)

Batch Processing ​

python
# Process multiple requests efficiently
import asyncio
from deepseek import AsyncDeepSeek

async def process_batch(prompts):
    client = AsyncDeepSeek(api_key="your-api-key")
    
    tasks = []
    for prompt in prompts:
        task = client.chat.completions.create(
            model="deepseek-chat",
            messages=[{"role": "user", "content": prompt}]
        )
        tasks.append(task)
    
    results = await asyncio.gather(*tasks)
    return results

# Usage
prompts = ["Hello", "How are you?", "Goodbye"]
results = asyncio.run(process_batch(prompts))

Knowledge Base ​

Frequently Asked Questions ​

General Questions ​

Q: How do I get started with DeepSeek? A: Follow our Quick Start Guide to create an account, get your API key, and make your first request.

Q: What programming languages are supported? A: We provide official SDKs for Python, JavaScript/TypeScript, Go, Java, C#, Ruby, and PHP. See our SDK documentation.

Q: How is pricing calculated? A: DeepSeek uses token-based pricing. You pay for input and output tokens consumed. See our pricing page for details.

Technical Questions ​

Q: What's the maximum context length? A: Context length varies by model:

  • DeepSeek Chat: 128K tokens
  • DeepSeek Coder: 128K tokens
  • DeepSeek Research: 256K tokens
  • DeepSeek Pro: 512K tokens

Q: How do I handle rate limits? A: Implement exponential backoff, reduce request frequency, or upgrade to higher limits. See our rate limiting guide.

Q: Can I fine-tune models? A: Custom fine-tuning is available for enterprise customers. Contact our sales team for details.

Billing Questions ​

Q: How am I billed? A: Billing is monthly based on token usage. You receive an invoice with detailed usage breakdown.

Q: Can I set spending limits? A: Yes, you can set monthly budgets and receive alerts in your dashboard.

Q: Do you offer refunds? A: Refunds are available for unused credits within 30 days, subject to terms of service.

Tutorials & Guides ​

Video Tutorials ​

Written Guides ​

Sample Applications ​

Contact Information ​

Support Contacts ​

General Support ​

Technical Support ​

Billing Support ​

Enterprise Support ​

  • Email: enterprise@deepseek.com
  • Phone: +1 (555) 123-4567
  • Response Time: 4 hours
  • Best For: Enterprise customers, custom solutions

Emergency Contacts ​

Critical Issues (Enterprise Only) ​

Office Locations ​

Headquarters ​

DeepSeek AI
123 Innovation Drive
San Francisco, CA 94105
United States
Phone: +1 (555) 123-4567

Regional Offices ​

Europe Office
456 Tech Street
London, UK EC2A 4DP
Phone: +44 20 1234 5678

Asia Pacific Office
789 AI Boulevard
Singapore 018956
Phone: +65 6123 4567

Service Level Agreements ​

Response Time Commitments ​

Free Tier ​

  • Email Support: 48-72 hours
  • Community Support: Best effort
  • Documentation: Self-service
  • Critical Issues: 4 hours
  • High Priority: 8 hours
  • Standard Issues: 24 hours
  • Low Priority: 48 hours

Enterprise ​

  • Critical Issues: 1 hour
  • High Priority: 2 hours
  • Standard Issues: 4 hours
  • Low Priority: 8 hours

Escalation Process ​

Automatic Escalation ​

  1. Initial Response: Support agent responds within SLA
  2. First Escalation: If no resolution within 2x SLA, escalate to senior agent
  3. Management Escalation: If no resolution within 4x SLA, escalate to management
  4. Executive Escalation: Critical issues escalated to executive team

Manual Escalation ​

  • Email: escalation@deepseek.com
  • Subject: Include "ESCALATION" and ticket number
  • Information: Provide ticket number, issue description, and business impact

Feedback & Improvement ​

Feedback Channels ​

Product Feedback ​

Support Feedback ​

  • Support Survey: Sent after each support interaction
  • Support Feedback: support-feedback@deepseek.com
  • Quarterly Reviews: Regular feedback sessions for enterprise customers

Continuous Improvement ​

Support Metrics ​

  • First Response Time: Average time to first response
  • Resolution Time: Average time to resolve issues
  • Customer Satisfaction: CSAT scores and feedback
  • Self-Service Success: Documentation effectiveness

Regular Reviews ​

  • Monthly: Support team performance review
  • Quarterly: Support process improvement
  • Annually: Comprehensive support strategy review

We're committed to providing exceptional support and continuously improving our services based on your feedback. Don't hesitate to reach out whenever you need assistance!

基于 DeepSeek AI 大模型技术

Copyright © 2025 DeepSeek | sitemap