Troubleshooting

Common issues and solutions for GMTech users. This guide covers everything from login problems to API errors, with step-by-step resolution instructions.

Account and Access Issues

Cannot Log In to GMTech

Problem: Login page shows "Invalid credentials"

Possible Causes:

• Incorrect email or password
• Account not yet activated
• Password reset needed
• Browser issues or cached data

Solutions:

Step 1: Verify Credentials

1. Double-check email address spelling
2. Ensure caps lock is off when typing password
3. Try typing password in a text editor first to verify
4. Check if you used social login (Google) instead of password

Step 2: Password Reset

1. Click "Forgot Password" on login page
2. Enter your email address
3. Check email (including spam folder) for reset link
4. Follow instructions to create new password
5. Try logging in with new password

Step 3: Browser Troubleshooting

1. Clear browser cache and cookies
2. Disable browser extensions temporarily
3. Try incognito/private browsing mode
4. Test with different browser (Chrome, Firefox, Safari)

Step 4: Account Activation

1. Check email for activation message after signup
2. Look in spam/junk folder for activation email
3. Request new activation email if needed
4. Contact support if activation email never arrives

Problem: "Account suspended" or "Access denied"

Possible Causes:

• Payment issues or failed billing
• Terms of service violations
• Unusual usage patterns flagged by security
• Team removal by admin

Solutions:

1. Check Payment Status: Verify your billing information and payment methods
2. Contact Team Admin: If you're a team member, check with team owner/admin
3. Review Terms: Ensure you haven't violated GMTech terms of service
4. Contact Support: Submit ticket with account details for investigation

Team Access Problems

Problem: Cannot access team workspace

Symptoms:

• 404 error when accessing team URL
• "Access denied" message
• Missing team in dashboard

Troubleshooting Steps:

Step 1: Verify Team URL

Correct Format: https://app.gmtech.com/a/team-name/
Common Errors:
❌ https://app.gmtech.com/team-name/
❌ https://app.gmtech.com/a/Team Name/ (spaces)
❌ https://app.gmtech.com/a/team_name/ (underscores)

Step 2: Check Team Membership

1. Go to main dashboard (https://app.gmtech.com)
2. Look for team in sidebar or team switcher
3. Verify you accepted team invitation
4. Check email for invitation that may need acceptance

Step 3: Contact Team Admin

1. Email team owner or admin to verify your status
2. Ask them to check team member list
3. Request new invitation if needed
4. Verify they have correct email address

AI Generation Issues

Model Not Responding

Problem: AI request hangs or times out

Possible Causes:

• Model temporarily unavailable
• Request too complex or long
• Network connectivity issues
• Rate limit exceeded

Solutions:

Step 1: Check Model Status

1. Try different model (switch from GPT-4 to Claude)
2. Refresh page and try again
3. Check if specific model is having issues

Step 2: Simplify Request

1. Reduce prompt length if very long
2. Break complex requests into smaller parts
3. Remove unnecessary context or examples
4. Try more specific, focused questions

Step 3: Network Troubleshooting

1. Check internet connection stability
2. Try different network if possible
3. Disable VPN temporarily
4. Clear browser cache and refresh

Step 4: Rate Limit Check

1. Wait 1-2 minutes before retrying
2. Check if you've made many recent requests
3. Verify team hasn't hit daily limits
4. Consider upgrading plan if hitting limits frequently

Problem: Error message appears instead of AI response

Common Error Messages and Solutions:

"Model currently unavailable"

• Cause: Provider maintenance or outage
• Solution: Try different model or wait 10-15 minutes

"Cost limit exceeded"

• Cause: Request would cost more than set limit
• Solution: Increase cost limit or use cheaper model

"Rate limit exceeded"

• Cause: Too many requests in short time
• Solution: Wait and retry, or upgrade plan

"Invalid request format"

• Cause: Malformed prompt or parameters
• Solution: Simplify prompt, check for special characters

"Insufficient credits"

• Cause: Account balance too low
• Solution: Add credits to account

Quality Issues

Problem: AI responses are poor quality or irrelevant

Diagnosis Questions:

• Is the prompt clear and specific?
• Are you using the right model for the task?
• Is there enough context provided?
• Are instructions too vague or complex?

Solutions:

Step 1: Improve Prompt Quality

❌ Poor Prompt:
"Write something about marketing"

✅ Improved Prompt:
"Write a 200-word email subject line for B2B software launch targeting CTOs. Focus on efficiency and ROI benefits."

Key Improvements:
├── Specific format (email subject)
├── Clear audience (CTOs)
├── Defined length (200 words)
├── Focused benefits (efficiency, ROI)

Step 2: Model Selection Optimization

Task-Model Matching:
├── Creative Writing → Claude-3.5-Sonnet
├── Technical Analysis → GPT-4 Turbo
├── Quick Questions → GPT-3.5-Turbo
├── Research Tasks → Gemini-1.5-Pro
├── Code Generation → GPT-4 Turbo or CodeLlama

Step 3: Context Enhancement

1. Provide relevant background information
2. Include examples of desired output format
3. Specify tone, audience, and constraints
4. Use system prompts for consistent behavior

Step 4: Iterative Refinement

1. Start with basic prompt
2. Add specific requirements based on initial response
3. Use follow-up questions to refine output
4. Save successful prompt patterns for reuse

API Integration Issues

Authentication Problems

Problem: "Invalid API key" errors

Verification Steps:

Step 1: Check API Key Format

# Correct format
GMTECH_API_KEY="gmtech_1234567890abcdef..."

# Common mistakes:
❌ Missing "gmtech_" prefix
❌ Extra spaces or characters
❌ Wrong environment variable name
❌ Quotes in wrong places

Step 2: Verify Key Status

1. Log into GMTech dashboard
2. Go to Settings → API Keys
3. Verify key is listed and active
4. Check key hasn't expired or been revoked
5. Regenerate key if necessary

Step 3: Test with cURL

curl -X GET "https://app.gmtech.com/api/v1/models" \
  -H "Authorization: Bearer $GMTECH_API_KEY" \
  -H "Content-Type: application/json"

Expected success response:

{
  "object": "list",
  "data": [
    {"id": "gpt-4-turbo", "object": "model"},
    {"id": "claude-3-5-sonnet-20241022", "object": "model"}
  ]
}

Problem: API requests fail intermittently

Possible Causes:

• Network connectivity issues
• Rate limiting
• Server-side temporary issues
• Client-side timeout settings

Solutions:

Step 1: Implement Retry Logic

import time
import random
from openai import APIError

def robust_api_call(client, **kwargs):
    max_retries = 3
    base_delay = 1

    for attempt in range(max_retries):
        try:
            return client.chat.completions.create(**kwargs)
        except APIError as e:
            if attempt == max_retries - 1:
                raise

            delay = base_delay * (2 ** attempt) + random.uniform(0, 1)
            time.sleep(delay)

Step 2: Monitor Rate Limits

# Check response headers for rate limit info
response = client.chat.completions.create(...)
print(f"Remaining requests: {response.headers.get('x-ratelimit-remaining')}")

Step 3: Increase Timeouts

client = OpenAI(
    base_url="https://app.gmtech.com/api/v1",
    api_key=api_key,
    timeout=60  # Increase from default 30s
)

Response Format Issues

Problem: Unexpected response format or missing fields

Common Issues:

Missing GMTech-specific fields

# Some responses may not have GMTech extensions
response = client.chat.completions.create(...)

# Safe access to GMTech fields
cost = getattr(response, 'gmtech_cost', 0)
provider = getattr(response, 'gmtech_provider', 'unknown')
response_time = getattr(response, 'gmtech_response_time', 0)

Streaming response issues

# Handle streaming chunks safely
stream = client.chat.completions.create(stream=True, ...)

for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        content = chunk.choices[0].delta.content
        print(content, end='', flush=True)

Model-specific response variations

# Different models may have slightly different response formats
try:
    content = response.choices[0].message.content
except (IndexError, AttributeError):
    print("Unexpected response format:", response)
    content = "Error: Could not extract content"

Billing and Payment Issues

Payment Problems

Problem: Payment declined or failed

Immediate Steps:

1. Check Card Details: Verify expiration date, CVV, billing address
2. Contact Bank: Ensure no holds on online/international transactions
3. Try Alternative Payment: Use different card or PayPal if available
4. Check Account Limits: Ensure sufficient credit limit or funds

Prevention:

1. Keep payment methods up to date
2. Set up backup payment methods
3. Enable automatic billing notifications
4. Monitor credit utilization

Problem: Unexpected charges or billing discrepancies

Investigation Steps:

Step 1: Review Usage Analytics

1. Go to Settings → Billing → Usage Analytics
2. Review detailed usage breakdown by date and team member
3. Look for unusual spikes in activity or costs
4. Check if API usage contributed to charges

Step 2: Check Team Activity

1. Review team member list for unauthorized access
2. Check API key usage logs
3. Look for automated scripts or integrations
4. Verify all team members are legitimate

Step 3: Analyze Model Usage

1. Check if expensive models were used unintentionally
2. Look for very long conversations or large prompts
3. Review image generation usage (higher cost per request)
4. Check for batch processing or bulk operations

Step 4: Contact Billing Support

1. Email billing@gmtech.com with specific details
2. Include invoice numbers and transaction dates
3. Provide screenshots of discrepancies
4. Request detailed usage breakdown if needed

Credit and Subscription Issues

Problem: Credits not appearing after purchase

Wait Period: Credits can take up to 24 hours to appear

Troubleshooting:

1. Check Payment Status: Verify payment was processed successfully
2. Clear Browser Cache: Refresh dashboard after clearing cache
3. Check Email: Look for payment confirmation email
4. Wait for Processing: Payment processing can take several hours
5. Contact Support: If credits missing after 48 hours

Problem: Subscription not active or features missing

Verification Steps:

1. Check Subscription Status: Go to Settings → Subscription
2. Verify Payment: Ensure latest payment was successful
3. Check Plan Features: Confirm current plan includes expected features
4. Team vs Personal: Ensure you're in correct team workspace

Common Solutions:

1. Payment Update: Add valid payment method if expired
2. Plan Upgrade: Upgrade to plan with needed features
3. Team Access: Get added to team with appropriate plan
4. Account Reactivation: Resolve any account suspension issues

Performance Issues

Slow Response Times

Problem: AI responses taking too long

Optimization Strategies:

Step 1: Model Selection

Speed Optimization:
├── Fastest: GPT-3.5-Turbo, Claude-3-Haiku
├── Balanced: Gemini-1.5-Pro, Command-R+
├── Quality-focused: GPT-4 Turbo, Claude-3.5-Sonnet

Use Case Matching:
├── Real-time chat: Use fastest models
├── Batch processing: Optimize for cost, not speed
├── Critical analysis: Accept slower response for quality

Step 2: Request Optimization

1. Shorter Prompts: Reduce input length when possible
2. Specific Requests: Avoid open-ended questions
3. Token Limits: Set appropriate max_tokens limits
4. Streaming: Use streaming for long responses

Step 3: Technical Optimization

1. Network: Check internet connection stability
2. Browser: Close other tabs, clear cache
3. Location: Use CDN-optimized endpoints when available
4. Time of Day: Consider peak usage times

Browser and Interface Issues

Problem: Interface not loading or behaving incorrectly

Browser Compatibility:

• Supported: Chrome 90+, Firefox 88+, Safari 14+, Edge 90+
• Not Supported: Internet Explorer, very old browser versions

Troubleshooting Steps:

Step 1: Basic Browser Fixes

1. Hard Refresh: Ctrl+F5 (Windows) or Cmd+Shift+R (Mac)
2. Clear Cache: Clear browser cache and cookies for app.gmtech.com
3. Disable Extensions: Test with extensions disabled
4. Incognito Mode: Try in private/incognito browsing

Step 2: JavaScript and Cookies

1. Enable JavaScript: Ensure JavaScript is enabled
2. Allow Cookies: Enable cookies for app.gmtech.com
3. Check Blockers: Disable ad blockers temporarily
4. Security Software: Check if antivirus/firewall is blocking

Step 3: Network and Connectivity

1. VPN Issues: Disable VPN and test
2. Corporate Firewall: Check if workplace firewall blocks app
3. DNS Issues: Try different DNS servers (8.8.8.8, 1.1.1.1)
4. Mobile vs Desktop: Test on different device types

Data and Conversation Issues

Conversation History Problems

Problem: Conversations disappearing or not saving

Possible Causes:

• Browser storage issues
• Account synchronization problems
• Session timeout
• Privacy settings

Solutions:

Step 1: Check Browser Settings

1. Local Storage: Ensure browser allows local storage
2. Cookies: Verify cookies are enabled for app.gmtech.com
3. Private Browsing: Regular mode saves more reliably than private
4. Storage Space: Clear browser storage if full

Step 2: Account Synchronization

1. Logout/Login: Sign out and back in to refresh session
2. Different Browser: Test conversation saving in different browser
3. Multiple Tabs: Avoid having multiple GMTech tabs open
4. Session Timeout: Refresh page if session expires

Step 3: Manual Backup

1. Copy Important Conversations: Save critical text before closing
2. Create Snapshots: Use snapshot feature for important conversations
3. Export Data: Use export features when available
4. Document Externally: Keep backup of critical AI outputs

Snapshot and Sharing Issues

Problem: Cannot create or access snapshots

Permission Checks:

1. Team Membership: Verify you have snapshot creation permissions
2. Plan Features: Ensure your plan includes snapshot functionality
3. Role Limitations: Check if your team role allows snapshots

Technical Issues:

1. Browser Compatibility: Ensure browser supports all features
2. Network Connectivity: Verify stable internet connection
3. File Size: Check if conversation is too large for snapshot
4. Server Issues: Try again after a few minutes

Problem: Shared snapshots not visible to team members

Verification Steps:

1. Sharing Settings: Confirm snapshot is set to "Team" or "Public"
2. Team Membership: Verify recipients are in same team
3. Access Permissions: Check team member roles and permissions
4. URL Correctness: Ensure correct snapshot URL is shared

Getting Additional Help

Self-Help Resources

Documentation:

• Getting Started Guide - Basic platform introduction
• API Documentation - Complete API reference
• Best Practices - Optimization strategies
• FAQ - Frequently asked questions

Community Resources:

• User Forums: Community discussions and tips
• Status Page: Real-time system status updates
• Release Notes: Latest features and updates
• Video Tutorials: Step-by-step visual guides

Contacting Support

When to Contact Support

• Account access issues that self-help doesn't resolve
• Billing discrepancies or payment problems
• Technical issues affecting multiple team members
• Security concerns or suspicious activity
• Feature requests or bug reports

How to Contact Support

Email Support: support@gmtech.com

• Response Time: 24-48 hours for standard issues
• Include: Account email, team name, detailed issue description
• Attachments: Screenshots of errors when applicable

Dashboard Tickets:

1. Go to Settings → Support
2. Click "Create New Ticket"
3. Fill out issue details completely
4. Attach relevant screenshots or logs

Live Chat: Available for Pro and Enterprise customers

• Hours: Business hours (9 AM - 6 PM EST)
• Access: Through dashboard when signed in
• Best For: Urgent technical issues

Information to Include in Support Requests

Account Details:

• Team name and account email
• Approximate time when issue occurred
• Browser type and version
• Operating system

Issue Description:

• Detailed steps to reproduce problem
• Error messages (exact text)
• Expected vs actual behavior
• Screenshots or screen recordings

Technical Details:

• API key ID (not the actual key)
• Model names used
• Request/response examples
• Network configuration if relevant

Previous Troubleshooting:

• Steps already tried from this guide
• Other solutions attempted
• Temporary workarounds discovered

Emergency Contacts

Critical Issues:

• Security Incidents: security@gmtech.com
• Billing Emergencies: billing@gmtech.com
• API Outages: status@gmtech.com

Enterprise Customers:

• Dedicated Support: Available through account manager
• Priority Queue: Faster response times
• Phone Support: Direct line for critical issues

Previous: ← Best Practices | Next: FAQ →