Pro Plan Details
The Alactic AGI Pro Plan is designed for professional users, small teams, and businesses requiring reliable document processing with access to both GPT-4o mini and full GPT-4o models. This guide covers complete technical specifications, capabilities, and optimization strategies.
Plan Overview
Target Users:
- Professional consultants and analysts
- Small to medium businesses (10-100 documents/month)
- Development teams building production applications
- Agencies serving multiple clients
- Researchers requiring advanced AI analysis
Key Advantages Over Free Plan:
- Access to GPT-4o model (16.7x more powerful)
- 5x more PDF processing capacity (100 vs 20)
- 4x more URL scraping capacity (200 vs 50)
- 50x more storage (5 GB vs 100 MB)
- 2x more RAM (8 GB vs 4 GB)
- Better VM performance (consistent vs burst-based)
Technical Specifications
Compute Resources
Virtual Machine:
- SKU: Standard_D2s_v3 (Azure Dsv3-series)
- vCPUs: 2 (dedicated, not burst)
- RAM: 8 GB
- Disk: 128 GB Premium SSD
- Network: Enhanced (1 Gbps)
- Temp Storage: 16 GB SSD
Performance Characteristics:
- CPU Performance: Consistent 100% utilization available
- No CPU Credits: Always full performance (unlike Free Plan)
- Best For: Production workloads requiring consistent performance
- Sustained Processing: Can handle back-to-back document processing without slowdown
Benchmark Comparison:
| Workload | Free Plan (B2s) | Pro Plan (D2s_v3) | Improvement |
|---|---|---|---|
| Single PDF (10 pages) | 22 seconds | 18 seconds | 18% faster |
| Batch 50 PDFs | 25 minutes | 15 minutes | 40% faster |
| URL scraping (complex) | 35 seconds | 25 seconds | 29% faster |
| Concurrent processing | 2 jobs | 5 jobs | 2.5x throughput |
Storage Capacity
Document Storage:
- Capacity: 5 GB total (50x increase from Free)
- File Size Limit: 100 MB per PDF (2x increase)
- Retention: Unlimited duration
- Performance: Premium SSD (faster read/write)
Cosmos DB (Database):
- Tier: Serverless
- Request Units: Up to 5,000 RU/s burst
- Storage: 50 GB max
- Typical Usage: 5-10 GB with 100 docs/month
Estimated Capacity:
- ~2,500 average PDFs (2 MB each)
- ~5,000 small PDFs (1 MB each)
- ~500 large PDFs (10 MB each)
- ~100 very large PDFs (50 MB each)
Processing Limits
Monthly Quotas:
- PDFs: 100 documents per month (5x vs Free)
- URLs: 200 websites per month (4x vs Free)
- Total Processing: 300 documents/month combined
- Storage: 5 GB
Rate Limits:
- Concurrent Jobs: 5 at once (2.5x vs Free)
- Queue Size: 200 jobs max (4x vs Free)
- Batch Size: 25 PDFs or 50 URLs per batch
Processing Speed:
| Document Type | Processing Time | Notes |
|---|---|---|
| Single-page PDF | 12-18 seconds | GPT-4o mini |
| Single-page PDF | 15-22 seconds | GPT-4o |
| 10-page PDF | 35-60 seconds | GPT-4o mini |
| 10-page PDF | 45-75 seconds | GPT-4o |
| 50-page PDF | 2-4 minutes | Either model |
| URL (article) | 8-15 seconds | Fast sites |
| URL (complex) | 25-40 seconds | JavaScript-heavy |
Consistent Performance:
- No CPU burst credit depletion
- Same speed for 1st and 100th document
- Suitable for production workloads
AI Model Access
Available Models:
- Full access to Alactic GPT-4o
- Full access to Alactic GPT-4o mini
Alactic GPT-4o (Pro Plan Highlight):
- Context Window: 128,000 tokens
- Max Output: 16,384 tokens
- Input Cost: $2.50 per 1M tokens
- Output Cost: $10.00 per 1M tokens
- Strengths:
- Superior reasoning for complex documents
- Better handling of ambiguous or technical content
- Excellent for legal, financial, medical analysis
- Nuanced understanding of context
- Creative synthesis and reformulation
When to Use Each Model:
Use GPT-4o for:
- Legal contracts (nuanced clauses)
- Financial reports (complex analysis)
- Medical documents (critical accuracy)
- Research papers (deep understanding)
- Technical specifications (precise interpretation)
Use GPT-4o mini for:
- Simple invoices and receipts
- Straightforward articles
- Product descriptions
- Meeting notes
- Regular correspondence
See Model Selection Guide for detailed decision framework.
API Access
Full API Access:
- RESTful API endpoints
- Same authentication (deployment key)
- All processing endpoints available
- Webhook support for async processing
- Batch API for bulk operations
API Rate Limits:
- 20 requests per second (2x vs Free)
- 5,000 requests per hour (5x vs Free)
- Sufficient for most automation workflows
Example API Usage:
# Process with GPT-4o
curl -X POST https://your-vm-ip/api/v1/process \
-H "X-Deployment-Key: ak-xxxxx" \
-F "file=@contract.pdf" \
-F "model=gpt-4o" \
-F "analysis_depth=deep"
# Returns detailed analysis with GPT-4o quality
Cost Breakdown
Infrastructure Costs (Fixed Monthly)
Baseline: $147.00/month
| Resource | SKU/Type | Cost/Month | Cost/Day |
|---|---|---|---|
| Virtual Machine | Standard_D2s_v3 | $105.00 | $3.50 |
| VM Disk | 128GB Premium SSD | $19.20 | $0.64 |
| Cosmos DB | Serverless | $12.00 | $0.40 |
| Storage Account | Standard LRS | $8.00 | $0.27 |
| Networking | Standard | $2.80 | $0.09 |
| Total Infrastructure | $147.00 | $4.90 |
These costs are constant - you pay $147/month regardless of usage.
Variable Costs (Model Usage)
GPT-4o mini costs (same as Free Plan):
Example 10-page document:
- Input: 8,000 tokens × $0.150 / 1M = $0.0012
- Output: 500 tokens × $0.600 / 1M = $0.0003
- Total: $0.0015 per document
GPT-4o costs (new in Pro Plan):
Example 10-page document:
- Input: 8,000 tokens × $2.50 / 1M = $0.020
- Output: 500 tokens × $10.00 / 1M = $0.005
- Total: $0.025 per document
Cost Comparison:
- GPT-4o is 16.7x more expensive per document
- But often worth it for complex analysis
Monthly Cost Scenarios
Scenario 1: GPT-4o mini only (cost-optimized)
- 100 PDFs with GPT-4o mini
- 200 URLs with GPT-4o mini
- Total documents: 300
- Average tokens: 8,500 per document
Calculation:
Infrastructure: $147.00
Model costs: 300 docs × 8,500 tokens × $0.18/1M = $0.46
Total: $147.46/month
Cost per document: $0.49
Scenario 2: Mixed usage (balanced)
- 50 PDFs with GPT-4o (complex documents)
- 50 PDFs with GPT-4o mini (simple documents)
- 200 URLs with GPT-4o mini
Calculation:
Infrastructure: $147.00
GPT-4o: 50 × $0.025 = $1.25
GPT-4o mini: 250 × $0.0015 = $0.38
Total: $148.63/month
Cost per document: $0.50
Scenario 3: GPT-4o heavy (quality-focused)
- 100 PDFs with GPT-4o
- 100 URLs with GPT-4o
- 100 URLs with GPT-4o mini
Calculation:
Infrastructure: $147.00
GPT-4o: 200 × $0.025 = $5.00
GPT-4o mini: 100 × $0.0015 = $0.15
Total: $152.15/month
Cost per document: $0.51
Key Insight: Even with heavy GPT-4o usage, model costs remain low compared to infrastructure.
Total Cost of Ownership
First Month:
- Deployment: $0 (no setup fees)
- Infrastructure: $147
- Model usage (typical): ~$2
- Total: $149
Annual Cost:
- $147 × 12 = $1,764 infrastructure
- ~$24 model usage
- Total: ~$1,788/year
Cost per Document:
- Infrastructure: $147 / 300 = $0.49 per document
- Models (mini): $0.0015 per document
- Models (GPT-4o): $0.025 per document
- Total: $0.49 - $0.52 per document depending on model mix
Comparison to Alternatives:
| Service | Cost Model | Equivalent Pro Plan Cost |
|---|---|---|
| AWS Textract | $1.50/1000 pages | $0.015/page × 3,000 pages = $45 |
| Azure Form Recognizer | $1.50/1000 pages | $45 |
| Google Document AI | $1.50/1000 pages | $45 |
| Alactic Pro Plan | $147 flat + tokens | $149 all-inclusive |
For 300 documents (avg 10 pages = 3,000 pages):
- Competitors: ~$45-50 for extraction only (no AI analysis)
- Alactic Pro: $149 for extraction + GPT-4o analysis + storage + API
Feature Matrix
Included Features
- Access to GPT-4o model
- Access to GPT-4o mini model
- PDF upload and processing
- URL scraping and extraction
- Text extraction from documents
- AI-powered summarization
- Key information extraction
- Entity recognition (people, companies, dates, amounts)
- Sentiment analysis (with Deep Analysis)
- Batch processing (up to 25 PDFs or 50 URLs at once)
- Vector storage for semantic search
- Full API access with webhooks
- Dashboard web interface
- JSON, Markdown, and Plain Text output formats
- Quick Extract, Standard Analysis, and Deep Analysis modes
- Email notifications
- Usage statistics and cost tracking
- Deployment key authentication
- Email support (2-3 business days)
- Community support (GitHub, forums)
Not Included (Pro+ or Enterprise Required)
- IP whitelisting (Pro+ required)
- Custom domain names (Pro+ required)
- Multi-region deployment (Enterprise)
- SLA guarantees (Enterprise)
- Priority support (Enterprise)
- Custom model fine-tuning (Enterprise)
- Scheduled batch processing (Enterprise)
- Advanced monitoring dashboards (Enterprise)
- Dedicated infrastructure (Enterprise)
Use Cases
Well-Suited for Pro Plan
1. Professional Services
Legal Services:
- Contract review: 50 contracts/month
- Due diligence: Analyze 20-30 documents per deal
- Legal research: Extract key points from case law
Example workflow:
- Upload 25 contracts for M&A due diligence
- Use GPT-4o for detailed clause analysis
- Extract obligations, deadlines, risks
- Generate summary report for client
- Cost: $147 + ~$0.63 model fees = $147.63
Financial Analysis:
- Quarterly report analysis: 40 reports/quarter
- Investment research: 100 documents/month
- Compliance document review
Consulting:
- Client document analysis
- Research report synthesis
- Competitive intelligence gathering
2. Small Business Operations
Invoice Processing:
- Process 100 invoices/month
- Extract: vendor, amount, date, line items
- Use GPT-4o mini for cost efficiency
Workflow:
- Upload batch of 25 invoices weekly
- GPT-4o mini extracts structured data
- Export to CSV for accounting system
- Cost: $0.0015 × 100 = $0.15 model fees
Contract Management:
- 20-30 vendor contracts
- Annual renewals and reviews
- SLA monitoring
HR Document Processing:
- Resume screening: 50 resumes/month
- Employee document management
- Policy document analysis
3. Content Operations
Content Curation:
- Scrape 200 articles/month
- Summarize key points
- Generate content briefs for writers
Example:
- Monitor 50 competitor blogs weekly
- Extract main topics, insights
- Compile into weekly digest
- Cost: GPT-4o mini sufficient
Market Research:
- Analyze 100 industry reports monthly
- Track competitive landscape
- Extract trends and insights
SEO and Marketing:
- Competitor content analysis
- Keyword extraction
- Content gap identification
4. Development and Testing
Production Application Development:
- Building document processing features
- Testing with realistic volumes
- Pre-production load testing
Why Pro Plan for development:
- Consistent performance (unlike Free's burst)
- GPT-4o for testing quality requirements
- Sufficient quota for realistic testing
- Production-like infrastructure
5. Agency Services
Serving 5-10 Clients:
- 10-20 documents per client per month
- Mixed use cases (legal, financial, content)
- White-label processing
Cost allocation:
Infrastructure: $147/month
Per client: $147 / 10 = $14.70/month
Mark up: Charge $50-100/client/month
Profit: $35-85 per client
Total revenue: $500-1000/month
Profit: $353-853/month
Not Suitable for Pro Plan
High-Volume Processing:
- More than 300 documents/month consistently
- Real-time processing for many users
- Batch processing of 1000+ documents
Better fit: Pro+ Plan (500 PDFs, 1,000 URLs)
Mission-Critical Production:
- Requires 99.9% uptime SLA
- Need 24/7 priority support
- Compliance requirements (SOC 2, HIPAA)
Better fit: Enterprise Plan
Very Large Documents:
- Regularly processing 100-500 page PDFs
- Need more than 100 MB per file
- Require extensive storage (more than 5 GB)
Better fit: Pro+ or Enterprise
Optimization Strategies
Maximize Value
1. Optimize Model Selection
Use GPT-4o strategically:
- Reserve GPT-4o for documents requiring deep analysis
- Use GPT-4o mini for straightforward extraction
- Implement cascading model strategy (try mini first, escalate if needed)
Example split:
- 80 documents: GPT-4o mini ($0.12)
- 20 documents: GPT-4o (complex) ($0.50)
- Total model cost: $0.62 vs $2.50 if all GPT-4o
2. Batch Processing Efficiency
Process in batches of 25 PDFs:
- Reduces overhead
- Faster total processing time
- Better resource utilization
3. Storage Management
Download and delete old results:
- Free up storage for new documents
- Archive externally (S3, Azure Blob, local)
- Keep only recent/active processing results
Target: Keep storage under 4 GB (buffer for spikes)
4. Quota Planning
Track usage throughout month:
- Week 1: Use 25% of quota (25 PDFs, 50 URLs)
- Week 2: Use 25% of quota
- Week 3: Use 25% of quota
- Week 4: Use 25% of quota + buffer
Prevents running out before month end.
5. API Integration
Automate routine processing:
- Weekly invoice batch processing (API scheduled job)
- Daily competitor content scraping
- Monthly report analysis
Reduces manual dashboard work.
Performance Tuning
Optimize Processing Speed
1. Use Quick Extract When Appropriate
- 3x faster than Standard Analysis
- Sufficient for simple extraction tasks
- Same text extraction quality
2. Disable Vector Storage If Not Needed
- Saves 2-3 seconds per document
- Only enable if you'll use semantic search
3. Choose Optimal Analysis Depth
| Depth | Speed | Use When |
|---|---|---|
| Quick Extract | Fastest | Only need text, no analysis |
| Standard Analysis | Balanced | Need summary + key points |
| Deep Analysis | Slowest | Need entities, sentiment, full analysis |
4. Process During Off-Peak Hours
- Azure OpenAI less congested
- Faster model inference
- Better for large batches
Optimize Costs
Cost Reduction Strategies:
1. Default to GPT-4o mini
- Set as default model in Settings
- Override to GPT-4o only when needed
- Saves 94% on model costs
2. Use Standard Analysis by Default
- Deep Analysis uses 2-3x more output tokens
- Reserve for important documents
3. Minimize Reprocessing
- Get settings right first time
- Avoid trial-and-error reprocessing
4. Compress Large PDFs Before Upload
- Reduces storage usage
- Faster upload and processing
- Use PDF compression tools
Monitoring and Management
Usage Dashboard
Track Key Metrics:
-
Documents Processed
- PDFs: X / 100
- URLs: Y / 200
- Percentage of quota used
-
Storage Usage
- Current: X GB / 5 GB
- Growth rate: MB per week
-
Cost Tracking
- Infrastructure: $147 (fixed)
- Model usage: $X (variable)
- Total month-to-date: $147 + $X
-
Model Distribution
- GPT-4o: X documents ($Y)
- GPT-4o mini: Z documents ($W)
- Cost per document average
Alerts and Notifications
Recommended Alert Thresholds:
- 75% quota used: "Plan remaining usage carefully"
- 90% quota used: "Consider upgrading or wait until next month"
- 100% quota: "Quota exceeded, upgrade or wait"
- 4 GB storage: "Clean up old documents"
- 4.5 GB storage: "Storage nearly full"
Set up in Settings → Notifications
Upgrading from Free Plan
Migration Process
Before upgrading:
-
Evaluate if Pro is right fit
- Need more than 70 documents/month?
- Need GPT-4o model quality?
- Building production application?
- If yes to any: Upgrade to Pro
-
Backup Free Plan data
- Download all results
- Export document metadata
- Document API integrations
-
Plan migration
- Schedule during low-activity period
- Update team on new deployment key
- Budget for higher monthly cost (+$75/month)
Migration steps:
- Delete Free Plan deployment
- Deploy Pro Plan from Azure Marketplace
- Update API integrations with new key
- Upload backed-up data if needed
- Verify all functionality
Downtime: 30-60 minutes
Cost Impact:
- Free: $72/month
- Pro: $147/month
- Increase: +$75/month (+104%)
When to Upgrade
Upgrade to Pro when:
- Consistently hitting Free Plan quotas
- Need GPT-4o for better analysis
- Require consistent performance (not burst-based)
- Building production application
- Need more storage (more than 100 MB)
Cost-benefit calculation:
If Free Plan insufficient:
- Cost of manual work or inferior results: $X
- Pro Plan additional cost: $75/month
- If X more than $75/month → Upgrade to Pro
Upgrading to Pro+
When to Consider Pro+
Upgrade to Pro+ when:
- Processing more than 300 documents/month regularly
- Need IP whitelisting for security
- Require more storage (more than 5 GB)
- Need faster processing (4 vCPUs vs 2)
- Production workload with higher scale
Pro+ Advantages:
- 500 PDFs + 1,000 URLs per month (vs 100 + 200)
- 20 GB storage (vs 5 GB)
- 4 vCPUs and 16 GB RAM (vs 2 vCPUs and 8 GB)
- IP whitelisting
- Custom domain support
Cost: $295/month (+$148/month vs Pro)
Break-even analysis:
If processing 400+ documents/month:
- With Pro: Limited by quota, can't process all
- With Pro+: Can process all documents
- Value of processing extra documents more than $148/month → Upgrade
Frequently Asked Questions
Q: Can I switch models mid-month without extra charge?
A: Yes. You can use GPT-4o and GPT-4o mini interchangeably. You only pay for actual token usage.
Q: What happens if I exceed 100 PDFs or 200 URLs?
A: Processing stops. You must wait until next month (quota resets 1st) or upgrade to Pro+ Plan.
Q: Can I share Pro Plan with my team?
A: Yes, multiple team members can use same deployment key. All usage counts against shared quota.
Q: How does Pro Plan compare to ChatGPT Plus for document analysis?
A: Pro Plan offers batch processing, API access, dedicated infrastructure, and specialized document training. ChatGPT Plus is for interactive chat only.
Q: Can I process documents larger than 100 MB?
A: No. Maximum file size on Pro Plan is 100 MB. For larger files, upgrade to Enterprise.
Q: Is Pro Plan suitable for production applications?
A: Yes, for moderate scale (up to 300 docs/month). For higher scale or SLA requirements, consider Pro+ or Enterprise.
Q: Can I pause Pro Plan to save costs?
A: You can deallocate the VM to save ~$105/month, but other resources still incur costs. Not recommended for production use.
Q: How do I estimate my monthly model costs?
A: Use formula: (Documents × Avg tokens × Model price per 1M tokens). For GPT-4o mini: typically $0.0015/document. For GPT-4o: typically $0.025/document.