8.6 KiB
8.6 KiB
Subscription Documentation Update Plan
Current State Analysis
Issues Found
-
Pricing Page Discrepancies:
- Documentation shows outdated plan limits
- Missing unified
ai_text_generation_calls_limitfor Basic plan (10 calls) - Missing video generation limits and pricing
- Missing Exa search pricing details
- Gemini pricing is outdated (docs show old models, code has 2.5 Pro, 2.5 Flash, etc.)
- Missing detailed Gemini model breakdowns
-
Missing Billing Dashboard Documentation:
- No documentation for dedicated billing dashboard page (
/billing) - Multiple dashboard components exist (BillingDashboard, EnhancedBillingDashboard, CompactBillingDashboard)
- No documentation for billing page features and usage
- No documentation for dedicated billing dashboard page (
-
Outdated Implementation Status:
- Documentation doesn't reflect current billing dashboard implementation
- Missing information about subscription renewal history
- Missing usage logs table documentation
- Missing comprehensive API breakdown component
Actual Values from Code
Subscription Plans (from pricing_service.py)
Free Tier
- Price: $0/month, $0/year
- Gemini calls: 100/month
- OpenAI calls: 0
- Anthropic calls: 0
- Mistral calls: 50/month
- Tavily calls: 20/month
- Serper calls: 20/month
- Metaphor calls: 10/month
- Firecrawl calls: 10/month
- Stability calls: 5/month
- Exa calls: 100/month
- Video calls: 0
- Gemini tokens: 100,000/month
- Monthly cost limit: $0.0
- Features: ["basic_content_generation", "limited_research"]
Basic Tier
- Price: $29/month, $290/year
- ai_text_generation_calls_limit: 10 (unified limit for all LLM providers)
- Gemini calls: 1000/month (legacy, not used for enforcement)
- OpenAI calls: 500/month (legacy)
- Anthropic calls: 200/month (legacy)
- Mistral calls: 500/month (legacy)
- Tavily calls: 200/month
- Serper calls: 200/month
- Metaphor calls: 100/month
- Firecrawl calls: 100/month
- Stability calls: 5/month
- Exa calls: 500/month
- Video calls: 20/month
- Gemini tokens: 20,000/month (increased from 5,000)
- OpenAI tokens: 20,000/month
- Anthropic tokens: 20,000/month
- Mistral tokens: 20,000/month
- Monthly cost limit: $50.0
- Features: ["full_content_generation", "advanced_research", "basic_analytics"]
Pro Tier
- Price: $79/month, $790/year
- Gemini calls: 5000/month
- OpenAI calls: 2500/month
- Anthropic calls: 1000/month
- Mistral calls: 2500/month
- Tavily calls: 1000/month
- Serper calls: 1000/month
- Metaphor calls: 500/month
- Firecrawl calls: 500/month
- Stability calls: 200/month
- Exa calls: 2000/month
- Video calls: 50/month
- Gemini tokens: 5,000,000/month
- OpenAI tokens: 2,500,000/month
- Anthropic tokens: 1,000,000/month
- Mistral tokens: 2,500,000/month
- Monthly cost limit: $150.0
- Features: ["unlimited_content_generation", "premium_research", "advanced_analytics", "priority_support"]
Enterprise Tier
- Price: $199/month, $1990/year
- All calls: Unlimited (0 = unlimited)
- All tokens: Unlimited (0 = unlimited)
- Video calls: Unlimited
- Monthly cost limit: $500.0
- Features: ["unlimited_everything", "white_label", "dedicated_support", "custom_integrations"]
API Pricing (from pricing_service.py)
Gemini API Models
- gemini-2.5-pro: $1.25/$10.00 per 1M input/output tokens
- gemini-2.5-pro-large: $2.50/$15.00 per 1M input/output tokens
- gemini-2.5-flash: $0.30/$2.50 per 1M input/output tokens
- gemini-2.5-flash-audio: $1.00/$2.50 per 1M input/output tokens
- gemini-2.5-flash-lite: $0.10/$0.40 per 1M input/output tokens
- gemini-2.5-flash-lite-audio: $0.30/$0.40 per 1M input/output tokens
- gemini-1.5-flash: $0.075/$0.30 per 1M input/output tokens
- gemini-1.5-flash-large: $0.15/$0.60 per 1M input/output tokens
- gemini-1.5-flash-8b: $0.0375/$0.15 per 1M input/output tokens
- gemini-1.5-flash-8b-large: $0.075/$0.30 per 1M input/output tokens
- gemini-1.5-pro: $1.25/$5.00 per 1M input/output tokens
- gemini-1.5-pro-large: $2.50/$10.00 per 1M input/output tokens
- gemini-embedding: $0.15 per 1M input tokens
- gemini-grounding-search: $35 per 1,000 requests (after free tier)
OpenAI Models
- gpt-4o: $2.50/$10.00 per 1M input/output tokens
- gpt-4o-mini: $0.15/$0.60 per 1M input/output tokens
Anthropic Models
- claude-3.5-sonnet: $3.00/$15.00 per 1M input/output tokens
HuggingFace/Mistral (GPT-OSS-120B via Groq)
- Configurable via env vars:
HUGGINGFACE_INPUT_TOKEN_COSTandHUGGINGFACE_OUTPUT_TOKEN_COST - Default: $1/$3 per 1M input/output tokens
Search APIs
- Tavily: $0.001 per search
- Serper: $0.001 per search
- Metaphor: $0.003 per search
- Exa: $0.005 per search (1-25 results)
- Firecrawl: $0.002 per page
Other APIs
- Stability AI: $0.04 per image
- Video Generation (HunyuanVideo): $0.10 per video generation
Billing Dashboard Components
Available Components
- BillingDashboard (
components/billing/BillingDashboard.tsx) - Main dashboard - EnhancedBillingDashboard (
components/billing/EnhancedBillingDashboard.tsx) - Enhanced version - CompactBillingDashboard (
components/billing/CompactBillingDashboard.tsx) - Compact version - BillingPage (
pages/BillingPage.tsx) - Dedicated billing page route
Features to Document
- Real-time usage monitoring
- Cost breakdown by provider
- Usage trends and projections
- System health indicators
- Usage alerts
- Subscription renewal history
- Usage logs table
- Comprehensive API breakdown
Update Plan
1. Update Pricing Page (docs-site/docs/features/subscription/pricing.md)
- Update all subscription plan limits to match actual database values
- Add unified
ai_text_generation_calls_limitexplanation for Basic plan - Update Gemini API pricing with all current models
- Update OpenAI pricing with actual values (gpt-4o, gpt-4o-mini)
- Update Anthropic pricing with actual values (claude-3.5-sonnet)
- Add Exa search pricing ($0.005 per search)
- Add video generation pricing and limits
- Add yearly pricing for all plans
- Update token limits to reflect actual values (20K for Basic, not 1M/500K)
- Add all search API limits per plan
- Add image generation limits per plan
- Add video generation limits per plan
2. Create/Update Billing Dashboard Documentation
- Create new page:
docs-site/docs/features/subscription/billing-dashboard.md - Document billing page route (
/billing) - Document all dashboard components (BillingDashboard, Enhanced, Compact)
- Document features: usage monitoring, cost breakdown, trends, alerts
- Document subscription renewal history component
- Document usage logs table
- Document comprehensive API breakdown component
- Add screenshots or descriptions of dashboard views
- Document how to access billing dashboard
3. Update Overview Page
- Add billing dashboard to features list
- Update supported API providers list (add Exa, Video generation)
- Update architecture to mention billing dashboard
4. Update Implementation Status
- Update to reflect billing dashboard implementation
- Add subscription renewal history feature
- Add usage logs table feature
- Update component count and features
5. Update API Reference
- Verify all endpoints are documented
- Add any missing endpoints for renewal history or usage logs
6. Update Navigation
- Add billing dashboard page to mkdocs.yml navigation
Priority Order
- High Priority: Update pricing page with correct values (users need accurate info)
- High Priority: Create billing dashboard documentation (major feature missing)
- Medium Priority: Update overview and implementation status
- Low Priority: Update API reference and navigation
Files to Update
docs-site/docs/features/subscription/pricing.md- Major update neededdocs-site/docs/features/subscription/overview.md- Minor updatesdocs-site/docs/features/subscription/implementation-status.md- Updates neededdocs-site/docs/features/subscription/billing-dashboard.md- NEW FILEdocs-site/mkdocs.yml- Add billing dashboard to nav
Notes
- The Basic plan has a critical unified limit:
ai_text_generation_calls_limit: 10- this applies to ALL LLM providers combined (Gemini, OpenAI, Anthropic, Mistral) - Token limits for Basic plan are much lower than documented: 20K per provider, not 1M/500K
- Video generation is a new feature with pricing and limits per plan
- Exa search is a separate provider from Metaphor with different pricing
- Multiple Gemini models exist with different pricing tiers
- Billing dashboard is a dedicated page, not just a component in main dashboard