Auction Pricing
This guide explains Hokusai's simple pricing model for model access.
Overview
Hokusai uses a straightforward pricing model that starts with minimal costs and can be adjusted through community governance. This approach ensures accessibility while allowing the community to respond to market conditions.
Running Example: Medical Image Analysis Model
Throughout this guide, we'll follow the pricing evolution of a medical image analysis model:
- Initial launch: Minimal price to cover costs
- Governance adjustments: Community-driven price changes
- Market response: Usage patterns and feedback
Pricing Model
Initial Price Setting
Models start with a minimal base price that covers only essential costs:
# Minimal cost components
compute_cost = 0.001 # Per inference
storage_cost = 0.0005 # Per image
network_cost = 0.0005 # Per request
initial_price = compute_cost + storage_cost + network_cost
# initial_price = 0.002 USDC per request
This low initial price ensures:
- Easy access for early adopters
- Minimal barriers to entry
- Market-driven usage patterns
Price Adjustments
Prices can be adjusted through community governance:
-
Proposal Process
- Any community member can propose a price change
- Proposals must include:
- Current price
- Proposed new price
- Justification for change
- Impact analysis
-
Voting
- Community members vote on proposals
- Majority approval required
- Minimum voting period: 7 days
- Minimum participation threshold: 10% of total tokens
-
Implementation
- Approved changes take effect after voting period
- Changes are implemented gradually over 30 days
- Users are notified of upcoming changes
Example: Price Adjustment Proposal
# Current price
current_price = 0.002 # USDC per request
# Proposed new price
proposed_price = 0.003 # 50% increase
# Implementation over 30 days
daily_increase = (proposed_price - current_price) / 30
# daily_increase = 0.000033 USDC
Usage Tiers
Basic Usage
- Pay-per-use pricing
- Standard queue priority
- Basic support
Batch Processing
- Volume-based discounts
- Priority queue access
- Enhanced support
Enterprise
- Custom pricing agreements
- Guaranteed capacity
- Dedicated support
- SLA guarantees
Example: Hospital Batch Processing
# Batch processing for hospital
batch_size = 1000 # Daily batch
current_price = 0.002 # Current price
volume_discount = 0.25 # 25% volume discount
batch_price = current_price * batch_size * (1 - volume_discount)
# batch_price = 0.002 * 1000 * 0.75 = 1.5 USDC
Best Practices
-
Cost Management
- Monitor usage patterns
- Use batch processing when possible
- Choose appropriate tier
-
Governance Participation
- Stay informed about price proposals
- Vote on important changes
- Provide feedback to the community
-
Budget Planning
- Set usage limits
- Monitor spending
- Plan for potential price changes
Troubleshooting
Common Issues
-
Price Changes
- Check governance proposals
- Review implementation schedule
- Plan for adjustments
-
Usage Optimization
- Use batch processing
- Monitor queue status
- Review tier options
-
Budget Management
- Set up alerts
- Monitor usage
- Review volume discounts
Next Steps
- Learn about Using Models
- Review Model API Guide
- Understand Tokenomics
For additional support, contact our Support Team or join our Community Forum.