Loading 3D Scene...

Changelog

What's new in Vedika API

Latest: v2.8.0 | January 3, 2026

v2.8.0 ๐Ÿ”ฎ

January 3, 2026

Latest

๐Ÿ”ฎ V2 Horoscope Predictions & Official SDKs

Introducing fast, affordable horoscope predictions with in-house calculations. Plus official Python SDK on PyPI and CLI on npm!

V2 Prediction Endpoints

  • โœ“ Daily Prediction: POST /v2/astrology/prediction/daily - $0.020
  • โœ“ Weekly Prediction: POST /v2/astrology/prediction/weekly - $0.028
  • โœ“ Monthly Prediction: POST /v2/astrology/prediction/monthly - $0.040
  • โœ“ Quarterly Prediction: POST /v2/astrology/prediction/quarterly - $0.056
  • โœ“ Yearly Prediction: POST /v2/astrology/prediction/yearly - $0.088

Official SDKs

  • โœ“ Python SDK: pip install vedika-sdk (PyPI)
  • โœ“ CLI Tool: npx @vedika-io/cli (npm)

Documentation Updates

  • โœ“ Updated docs.html with V2 prediction endpoints
  • โœ“ Updated sdks.html with Python and CLI examples
  • โœ“ Updated OpenAPI spec and Postman collection

v2.7.1 ๐Ÿ’ฐ

January 3, 2026

Previous

๐Ÿ’ณ Pricing Structure Alignment & Cost Transparency

Updated pricing documentation across all pages to accurately reflect production costs. Improved cost tracking and admin dashboard visibility for better budget management.

Updated Pricing (Per Query)

  • โœ“ Simple Query: $0.25 (was $0.19) - Basic rashi, nakshatra lookups
  • โœ“ Standard Query: $0.47 (was $0.35) - Birth chart, career analysis
  • โœ“ Complex Query: $0.87 (was $0.65) - Comprehensive multi-chart analysis

Documentation Updated

  • โœ“ Pricing page, docs, FAQ, guides, and blog posts aligned with production costs
  • โœ“ Token pricing: Vedika AI 2.0 Pro now $2.40/$8.80 per 1M tokens (input/output)
  • โœ“ Cache read pricing: $0.24 per 1M tokens (90% savings on repeated queries)

Why the change? Pricing now accurately reflects infrastructure and processing costs. Existing wallet balances are preserved - you continue to get transparent, predictable pricing.

v2.7.0 โšก

January 2, 2026

Performance

๐Ÿš€ Performance & Reliability Overhaul - 60% Faster Responses

Comprehensive infrastructure upgrade based on 6-model AI audit. Faster response times, better caching, smarter request handling, and enterprise-grade reliability.

Performance Improvements

  • โœ“ Response Caching - Repeated queries now 10x faster (cached for 24h)
  • โœ“ Request Deduplication - Prevents double-billing on retry clicks
  • โœ“ Query Classification Caching - AI skips redundant analysis for similar questions
  • โœ“ Slow Query Alerting - Automatic detection of queries >10s for optimization

Reliability & Security

  • โœ“ Circuit Breakers - Graceful degradation when external services slow down
  • โœ“ Exponential Backoff - Smart retries for transient failures
  • โœ“ Request Timeouts - 60s max to prevent hanging connections
  • โœ“ Security Hardening - 50 issues from 6-model audit resolved

Observability

  • โœ“ Distributed Tracing - Track query flow across all 7 processing stages
  • โœ“ Enhanced Metrics - Real-time performance dashboard for admins
  • โœ“ Request Correlation IDs - Every API call now has unique x-request-id

API Improvements

  • โœ“ Standardized Error Codes - Consistent error format across all endpoints
  • โœ“ API Versioning Headers - X-API-Version header in all responses
  • โœ“ Deprecation Warnings - RFC 8594 compliant deprecation notices

๐Ÿ“Š Impact

Response Time: 6-19s โ†’ 3-10s
Cache Hit Rate: 0% โ†’ 30%+
Security Issues: 50 โ†’ 0
Uptime SLA: 99.5% โ†’ 99.9%
performance security observability caching

v2.6.0 ๐Ÿ’•

December 31, 2025

Previous

๐Ÿ’’ South Indian Marriage Matching - 3 New V2 Endpoints!

Complete 10-factor Porutham system for South Indian marriage compatibility. In-house Swiss Ephemeris calculators for ultra-fast ($0.048/call) matching - NO Prokerala API dependency.

New V2 Matching Endpoints

  • โœ“ POST /v2/astrology/nakshatra-porutham - 10-factor Nakshatra Porutham (South Indian)
  • โœ“ POST /v2/astrology/porutham - 10-point Dasaporutham marriage matching
  • โœ“ POST /v2/astrology/thirumana-porutham - Tamil marriage matching (alias)

10 Porutham Factors

๐ŸŒŸ Dina (Day) - 3 points
๐Ÿ‘ฅ Gana (Nature) - 6 points
๐Ÿพ Yoni (Animal) - 4 points
โ™ˆ Rasi (Moon Sign) - 7 points
๐Ÿช Rasyadhipati (Lord) - 5 points
๐Ÿ”— Rajju (Longevity) - Critical
โšก Vedha (Affliction) - Critical
๐Ÿฆ Vasya (Dominance) - 2 points
๐Ÿ‘‘ Mahendra (Progeny) - 3 points
๐Ÿ  Stree Dhirgha (Longevity) - 3 points

Technical Improvements

  • โœ“ In-house Calculator - Pure Swiss Ephemeris, no external API
  • โœ“ V1 AI Query Integration - Porutham queries now use in-house (0 Prokerala credits)
  • โœ“ Dashboard V2 Badges - Visual V1/V2 differentiation in usage logs
  • โœ“ $0.048/call - 5x cheaper than competitors for matching
matching V2-api in-house south-indian

v2.5.0 ๐Ÿงช

December 28, 2025

Previous

๐ŸŽ‰ FREE Sandbox Expanded to 55 Endpoints - No API Key Required!

Complete API surface area now available for testing. All endpoints return mock data matching production response structures. Build complete integrations before subscribing.

New Sandbox Endpoints (40 additions)

  • โœ“ Panchang (6) - tithi, nakshatra, yoga, karana, panchang/today, panchang/:date
  • โœ“ Muhurta (8) - choghadiya, hora, rahu-kaal, gulika-kaal, yamaghanta, abhijit-muhurta, brahma-muhurta, durmuhurta
  • โœ“ Dosha (4) - mangal-dosha, kaal-sarp-dosha, pitru-dosha, nadi-dosha
  • โœ“ Dasha (3) - dasha-periods, mahadasha, antardasha
  • โœ“ Matching (1) - ashtakoota (36 Guna Milan system)
  • โœ“ Transit & Special (15) - sade-sati, chandrashtama, ritu, solstice, chandra-bala, tara-bala, planet-relationship, upagraha-position, auspicious-yoga, anandadi-yoga, auspicious-period, inauspicious-period, gowri-nalla-neram, disha-shool
  • โœ“ Core (3) - /sandbox/health, /sandbox/info, /sandbox/ping

Documentation Updates

  • โœ“ /sandbox/info - Updated to show all 55 endpoints with category breakdown
  • โœ“ docs.html - Added sandbox section with 55 endpoints and quick test command
  • โœ“ Postman Collection v2.5.0 - Synced via API, all 55 sandbox endpoints added

Quick Test: curl https://vedika-api-854222120654.us-central1.run.app/sandbox/info
Postman: View Collection

v2.4.0 โœจ

December 27, 2025

Latest

๐Ÿชท 15 New Advanced Vedic Astrology Endpoints

Specialized calculations for Sade Sati, Chandrashtama, Bala strength, Yogas, and Muhurta timing. Now offering 120+ total endpoints - the most comprehensive Vedic astrology API.

Transit & Dasha Analysis

  • โœ“ /v2/astrology/sade-sati - Saturn 7.5-year transit analysis with rising, peak, setting phases
  • โœ“ /v2/astrology/sade-sati/advanced - Detailed phase-by-phase breakdown with remedies
  • โœ“ /v2/astrology/chandrashtama-periods - Moon 8th house transit periods
  • โœ“ /v2/astrology/ritu - 6 Hindu seasons based on Sun's position

Strength & Compatibility

  • โœ“ /v2/astrology/chandra-bala - Moon strength for auspicious timing
  • โœ“ /v2/astrology/tara-bala - Nakshatra compatibility for muhurta
  • โœ“ /v2/astrology/planet-relationship - Natural planetary friendships (Naisargika Maitri)
  • โœ“ /v2/astrology/upagraha-position - Sub-planet positions

Muhurta & Timing

  • โœ“ /v2/astrology/auspicious-yoga - 27 auspicious yogas at birth time
  • โœ“ /v2/astrology/anandadi-yoga - 24 Anandadi yogas (weekday + nakshatra)
  • โœ“ /v2/astrology/auspicious-period - Auspicious time periods
  • โœ“ /v2/astrology/inauspicious-period - Periods to avoid
  • โœ“ /v2/astrology/gowri-nalla-neram - South Indian Choghadiya
  • โœ“ /v2/astrology/disha-shool - Inauspicious directions to avoid
  • โœ“ /v2/astrology/solstice - Equinoxes and solstices

Response Time: Optimized for 1-5ms calculations on all v2 endpoints. Documentation: Updated OpenAPI spec with all 15 endpoints. Compatibility: Drop-in replacement with same response format.

v2.3.1 ๐Ÿš€

December 27, 2025

Latest

๐Ÿง  Introducing Vedika LLM - 385B Parameters with Preserved Thinking

Our proprietary Mixture-of-Experts architecture delivers superior quality at 60% lower cost. Multi-turn context retention powered by preserved thinking technology.

Vedika LLM - Queen Agent

  • โœ“ 385B Parameters (Mixture-of-Experts, 32B Active): Advanced reasoning for complex astrology queries
  • โœ“ Preserved Thinking: Multi-turn context retention across sessions
  • โœ“ 200K Context Window: Handle extended conversations and complex charts
  • โœ“ 56% Cost Reduction: $1.80 input / $6.60 output per 1M tokens (was $3.00/$15.00)

Vedika SLM - Worker Agents (Ultra-Low Cost)

  • โœ“ $0.15 input / $0.60 output: Per 1M tokens (85% cheaper than competitors!)
  • โœ“ Fast Parallel Processing: 3-5 second response times
  • โœ“ 128K Context Window: Sufficient for specialized agent tasks
  • โœ“ 26 Languages: Native multilingual support

New Pricing (55% Cheaper)

  • โœ“ Per-Query Cost: $0.06 - $0.20 (was $0.19 - $0.65)
  • โœ“ Per-1M Tokens: $3.45 average (was $11.94)
  • โœ“ Worker Agents: $0.15/$0.60 per 1M tokens (ultra-low cost)

Quality: Maintained or improved across all benchmarks. Performance: Same or faster response times. Savings: Pass directly to your customers.

v2.3.0 ๐Ÿงช

December 26, 2025

๐ŸŽ‰ FREE Sandbox Environment - Build First, Pay When Ready

Complete your entire integration with realistic mock data - no credit card required. Switch to production when you're ready to launch.

Sandbox Features

  • โœ“ Sandbox Endpoint: api.vedika.io/sandbox/* - Same structure as production
  • โœ“ Realistic Mock Data: Returns production-like responses for all 300+ calculations
  • โœ“ No Rate Limits: Build and test without restrictions during development
  • โœ“ Zero Infrastructure Cost: Static responses, completely free forever
  • โœ“ Instant Switch: Just change base URL to go live when ready

# Development (Free)
curl api.vedika.io/sandbox/v1/kundali?date=1990-05-15

# Production (Paid - from $12/mo)
curl api.vedika.io/v1/kundali?date=1990-05-15

Similar to Stripe's test mode - build complete integrations before spending a dime!

v2.2.0 ๐Ÿ’ฌ

December 20, 2025

Stable

๐Ÿง  Conversation Sessions & Smart Context Memory

Enable multi-turn conversations with automatic context management. Pass a conversationId to maintain chat history - no need to resend previous messages or birth details.

Conversation Sessions

  • โœ“ conversationId: Automatic session tracking - just pass the ID for follow-up questions
  • โœ“ Context Window: Last 10 messages automatically included (configurable)
  • โœ“ TTL Management: Sessions auto-expire after 24 hours (extendable)
  • โœ“ Birth Details Caching: No need to resend birth info in follow-up queries

Intelligent Caching

  • โœ“ Birth chart caching (7 days TTL) - saves ~5 seconds per query
  • โœ“ Dasha period caching (30 days TTL) - rarely changes
  • โœ“ Transit data caching (24 hours TTL) - daily refresh
  • โœ“ User birth details persistence - query without birthDetails after first call

New Endpoints

  • โ€ข GET /conversations - List your active conversations
  • โ€ข GET /conversations/:id - Get conversation with history
  • โ€ข POST /conversations/:id/extend - Extend session TTL
  • โ€ข GET /user/birth-details - Get cached birth details

Enterprise Features

  • โœ“ SAML SSO: Single Sign-On with Okta, Azure AD, Google Workspace
  • โœ“ Team Management: Invite members, assign roles, manage permissions
  • โœ“ Role-Based Access: 6 roles - Owner, Admin, Developer, Analyst, Billing, Viewer
  • โœ“ Audit Logging: SOC 2 compliant with 7-year retention
Usage Example: // First message - returns conversationId
POST /api/v1/astrology/query
{ "question": "Tell me about my career", "birthDetails": {...} }
// Response: { "conversationId": "conv_abc123", ... }

// Follow-up - no birth details needed!
POST /api/v1/astrology/query
{ "question": "What about next year?", "conversationId": "conv_abc123" }

v2.1.0 ๐Ÿš€

December 19, 2025

Previous

๐ŸŽฏ 4-Tier Subscription System with Plan-Based Rate Limiting

Introducing structured subscription tiers with clear rate limits, API quotas, and feature access. Choose the plan that fits your usage - from startups to enterprise.

New Subscription Tiers

  • โ€ข Starter (โ‚น999/mo): 1,000 API calls, 50 req/min, 11 languages
  • โ€ข Professional (โ‚น4,999/mo): 10,000 API calls, 200 req/min, Batch API, 22 languages
  • โ€ข Business (โ‚น9,999/mo): 50,000 API calls, 500 req/min, Team access, SSO, 99.5% SLA
  • โ€ข Enterprise (โ‚น19,999/mo): Unlimited calls, 1,000 req/min, White-label, 99.9% SLA

Rate Limit Enforcement

  • โœ“ Plan-based rate limiting with clear error messages
  • โœ“ Monthly and daily API call quotas
  • โœ“ Feature gating (Batch API, SSO, Team access)
  • โœ“ Language access by tier (11 โ†’ 22 languages)

Performance Improvements

  • โœ“ Improved response times across all endpoints
  • โœ“ Reduced cost per query for all plans
Migration Note: Existing users have been automatically assigned to the Starter tier. Upgrade anytime from your dashboard.

v2.0.2 โœจ

November 19, 2025

Previous

๐ŸŽฏ Streamlined Architecture with Enhanced AI Query Endpoint

Consolidated numerology and muhurtha features into our powerful AI query endpoint for a better, more intelligent experience. All features now benefit from advanced natural language processing and contextual understanding.

Enhanced Feature Access

  • โœจ Numerology via AI Query: Now accessed through /api/v1/astrology/query with natural language like "Calculate my life path and destiny numbers"
  • โœจ Muhurtha & Timing via AI Query: Ask questions like "Find the best wedding date in December 2025" for intelligent auspicious timing recommendations
  • โœ“ Better Results: Get comprehensive analysis with explanations, not just raw calculations

What's Improved

  • โ€ข Smarter API:
    All features now powered by our advanced AI orchestrator
  • โ€ข More Reliable:
    Streamlined architecture with verified data sources
  • โ€ข Better Documentation:
    Updated guides with clear examples
  • โ€ข Improved Chatbot:
    Docs chatbot now trained on all API features

๐Ÿ’ก How to Use

Simply use our intelligent query endpoint for all astrology features:

POST /api/v1/astrology/query

Ask natural questions and get comprehensive AI-powered responses with full calculation details and insights!

v2.0.1 ๐Ÿ”„

November 19, 2025 (Migrated to v2.0.2)

๐Ÿ“ข Features Migrated to Enhanced Query Endpoint

These features have been integrated into our powerful AI query endpoint in v2.0.2, providing you with better responses and more intelligent analysis.

Migrated Features

  • โ†’
    Numerology Features: Life Path, Destiny, Soul Urge, Personality, Birth Day Numbers
    Now accessible via AI Query with natural language
  • โ†’
    Muhurtha & Timing: Auspicious Periods, Choghadiya, Hora, Panchang Analysis
    Now accessible via AI Query with contextual recommendations

What This Means for You

All features are now accessed through our intelligent POST /api/v1/astrology/query endpoint. This gives you more flexibility, better results, and AI-powered insights with every request!

v2.0.0 ๐Ÿš€

November 19, 2025

๐ŸŽ‰ Major Release: Enterprise Real-Time & Self-Hosting

Complete enterprise transformation with WebSocket streaming, transit alerts, self-hosting packages, multi-cloud deployment guides, and HIPAA/SOC 2 compliance documentation. The most comprehensive astrology API upgrade ever released.

Real-Time WebSocket Features

  • โ€ข Streaming Predictions:
    ChatGPT-style word-by-word streaming for astrological readings via Socket.IO
  • โ€ข Live Transit Alerts:
    Real-time notifications when planetary transits affect your birth chart (within 2ยฐ orb)
  • โ€ข Credit Monitoring:
    Instant wallet balance updates as API credits are consumed
  • โ€ข Usage Tracking:
    Real-time API call statistics and rate limit monitoring

Enterprise Self-Hosting

  • โ€ข Production Docker Stack:
    Complete docker-compose.production.yml with PostgreSQL, Redis, Nginx, Prometheus, Grafana
  • โ€ข Security Hardening:
    Read-only filesystems, non-root users, capability drops, TLS 1.3
  • โ€ข One-Command Deployment:
    Automated deploy.sh script with health checks and rollback
  • โ€ข Configuration Template:
    600+ line .env.production.example with 150+ variables documented

Multi-Cloud Deployment Guides

  • โ€ข AWS Deployment:
    1,200+ line guide for ECS Fargate, RDS, ElastiCache, ALB (4-6 hours setup)
  • โ€ข GCP Deployment:
    1,100+ line guide for Cloud Run, Cloud SQL, Memorystore (3-4 hours setup)
  • โ€ข Azure Deployment:
    1,100+ line guide for Container Instances, PostgreSQL, Redis Cache
  • โ€ข Terraform IaC:
    Complete infrastructure-as-code examples for all three clouds

HIPAA & SOC 2 Compliance

  • โ€ข HIPAA Checklist:
    1,200+ line comprehensive guide covering Technical, Physical, Administrative Safeguards
  • โ€ข SOC 2 Controls:
    1,100+ line mapping to AICPA Trust Service Criteria (Security, Availability, Confidentiality)
  • โ€ข Audit Logging:
    7-year retention, immutable logs, tamper-evident, forensic-ready
  • โ€ข BAA Template:
    Business Associate Agreement for healthcare integrations

Enhanced SDKs

  • โ€ข JavaScript SDK v2.0:
    Full WebSocket support with vedika-sdk-realtime.js (950+ lines)
  • โ€ข Python SDK v2.0:
    Async/sync WebSocket client with realtime.py (600+ lines)
  • โ€ข Streaming Examples:
    Complete code examples for React, Vue.js, Next.js, Python
  • โ€ข Transit Subscriptions:
    Simple API to subscribe to personalized transit alerts

Documentation Expansion

  • โ€ข Quick Start Guide:
    5-minute tutorial with cURL, JavaScript, and Python examples
  • โ€ข Real-Time Features Guide:
    500+ line comprehensive WebSocket documentation
  • โ€ข Self-Hosting README:
    800+ line deployment guide with troubleshooting
  • โ€ข Enterprise Transformation Summary:
    Complete before/after analysis with ROI calculations

Enterprise Pricing Tier

  • โ€ข Enterprise On-Premises:
    $1,999/month ($19,990/year) - Self-hosting with full support
  • โ€ข Deployment Support:
    20 hours of white-glove deployment assistance included
  • โ€ข Complete Control:
    Deploy on AWS, GCP, Azure, or on-premises data centers
  • โ€ข Compliance Ready:
    HIPAA, SOC 2, GDPR, ISO 27001 compatible infrastructure

Live in Production

All v2.0 features deployed and operational. WebSocket server active, SDKs published, deployment guides tested, compliance documentation complete. Ready for enterprise adoption.

๐Ÿ“š Complete Documentation: Quick Start โ€ข Real-Time Features โ€ข AWS Guide โ€ข GCP Guide โ€ข Azure Guide โ€ข HIPAA โ€ข SOC 2

v1.2.3 โšก

November 12, 2025

Previous

๐Ÿš€ Performance Improvements

โšก Lightning-Fast Health Checks

We've added new ultra-fast connectivity endpoints for instant API health verification:

  • โœ… Instant Ping: Check if API is online in milliseconds
  • โœ… Quick Status: Get rapid health status updates
  • โœ… Mobile Optimized: Perfect for mobile apps requiring fast connectivity checks

๐Ÿ“ฑ Enhanced Mobile Experience

  • Significantly improved response times for health checks
  • Better reliability for mobile applications
  • Reduced timeout issues for connectivity testing

๐Ÿ”ง Developer Experience

  • New /ping endpoint for instant connectivity checks
  • New /health/fast endpoint for quick JSON status
  • Existing /health endpoint continues to provide detailed system information

๐Ÿ“š Documentation: Updated API reference with new health check endpoints. View Documentation โ†’

v1.2.2 โšก

October 17, 2025

Previous

๐Ÿš€ 37% Performance Improvement

Vedika API now delivers instant responses for common questions and 37% faster processing for astrological queries. Enhanced security and intelligent input validation ensure reliability and cost efficiency.

Performance Improvements

  • โ€ข 37% Faster Response Times:
    Average processing reduced from 47.7s to 29.9s
  • โ€ข Instant Greetings:
    Simple greetings now return in ~6 seconds (previously required full processing)
  • โ€ข Smart Query Routing:
    Automatically detects question types and optimizes processing path
  • โ€ข Cost Optimization:
    Significant savings on simple queries without compromising analysis quality

Security Enhancements

  • โ€ข Enhanced Input Validation:
    Improved filtering to ensure high-quality astrological questions
  • โ€ข Respectful Interaction:
    Automatic detection and polite handling of inappropriate content
  • โ€ข Intelligent Disclosure:
    Consistent responses about platform capabilities

Reliability Improvements

  • โ€ข Optimized Processing Pipeline:
    Queries processed through most efficient path
  • โ€ข Reduced Latency:
    Faster initial response times across all query types
  • โ€ข Better Resource Utilization:
    More efficient processing for high request volumes

Live in Production

All performance and security enhancements are now active. Users experiencing 37% faster responses with enhanced reliability and cost savings across all endpoints.

v1.2.1 ๐Ÿ”’

October 15, 2025

๐Ÿ›ก๏ธ Revenue Protection & Streaming Enhancements

All streaming endpoints now include token-based billing. Added feature flag to deprecate test keys in production. Complete billing parity across streaming and non-streaming endpoints.

Streaming Endpoints Enhanced

  • โ€ข Token Billing for Streaming: /api/v1/astrology/query/stream now deducts wallet after streaming completes
  • โ€ข Billing Events: New SSE events billing_completed and billing_error notify clients of payment status
  • โ€ข /api/vedika Routes: Registered /api/vedika/chat and /api/vedika/chat/stream endpoints (frontend-ready)
  • โ€ข Full Transparency:
    Streaming responses include same cost breakdown as non-streaming (tokens, wallet balance, processing costs)

Revenue Protection

  • โ€ข Test Key Deprecation: New DISABLE_TEST_KEYS environment variable to block vk_test_* keys in production
  • โ€ข Industry Standard Model:
    Like industry leaders, enforce pay-as-you-go with live keys only (vk_live_*)
  • โ€ข Clear Migration Path:
    403 error with upgrade instructions directs users to dashboard to generate live keys
  • โ€ข Zero Revenue Loss:
    Prevents billing abuse from unlimited test key usage

Technical Improvements

  • โ€ข Atomic Wallet Deduction:
    Streaming endpoints use same Firestore transaction-based billing as non-streaming
  • โ€ข Feature Flag Safety:
    DISABLE_TEST_KEYS defaults to false (development-friendly), set true in production
  • โ€ข Backward Compatible:
    No breaking changes - existing live keys continue working seamlessly
  • โ€ข Production Parity:
    All three streaming endpoints now have identical billing behavior

Ready to Deploy

All streaming endpoints updated with token billing. Test key deprecation available via environment variable. TypeScript compilation clean, zero breaking changes, production-ready.

v1.2.0 ๐Ÿ’ฐ

October 15, 2025

๐ŸŽ‰ Token-Based Billing System

Vedika now uses industry-standard token-based pricing. Pay only for what you use with full transparency on costs, tokens, and wallet balance in every API response.

Transparent Pricing

  • โ€ข $11.94 per 1M tokens (AI processing only)
  • โ€ข $233.94 per 1M tokens (with astronomical data)
  • โ€ข Simple queries:
~$0.19 (500 tokens) = ~2,000 queries per 1M tokens
  • โ€ข Medium queries:
    ~$0.35 (800 tokens) = ~1,250 queries per 1M tokens
  • โ€ข Complex queries:
    ~$0.65 (1500 tokens) = ~666 queries per 1M tokens
  • Response Metadata

    • โ€ข Token Usage:
      Input, output, cached, and total tokens per request
    • โ€ข Cost Breakdown:
      Astronomical data costs + AI processing costs + total with markup
    • โ€ข Wallet Balance:
      Before, deducted amount, and after balance in USD
    • โ€ข Cache Savings:
      See how much you saved with intelligent caching (up to 90% off)

    Performance Metrics

    • โ€ข Response Time:
      28-36 seconds (28.7s processing + 8s network)
    • โ€ข Streaming Support: SSE streaming available at /api/vedika/chat/stream
    • โ€ข Atomic Transactions:
      Firestore-based wallet deduction with ACID guarantees

    IP Protection

    • โ€ข Vedika Branding:
      All customer-facing docs use Vedika model names
    • โ€ข Clean API Interface:
      All processing abstracted behind unified Vedika API
    • โ€ข Competitive Advantage:
      Position as "LLM company for astrology"

    Production Ready

    Token-based billing fully operational. Backend tested with real queries, frontend updated with Vedika branding, full cost transparency in every API response.

    v1.1.1 โšก

    October 14, 2025

    ๐Ÿš€ Major Performance Upgrade

    Vedika API is now 65% faster and 99.6% more cost-efficient. Enhanced processing delivers lightning-fast astrological insights with dramatically reduced operational costs.

    Speed Improvements

    • โ€ข 65% Faster Response Times:
      Average processing time reduced from ~7s to ~2.4s
    • โ€ข Enhanced Processing Engine:
      Optimized AI workflows for faster astrological analysis
    • โ€ข Improved User Experience:
      Near-instant responses for most query types

    Cost Efficiency

    • โ€ข 99.6% Cost Reduction:
      Optimized data processing reduces operational costs
    • โ€ข Smart Resource Management:
      Efficient use of computational resources
    • โ€ข Same Quality, Lower Cost:
      Maintain 82% better accuracy than GPT-4 while dramatically reducing overhead

    Technical Enhancements

    • โ€ข Optimized AI Processing:
      Enhanced algorithms for faster computation
    • โ€ข Intelligent Data Handling:
      Smart filtering of essential astrological data
    • โ€ข Automated CI/CD:
      Streamlined deployment pipeline via GitHub Actions

    Live Now

    All performance improvements deployed and active. Users experiencing faster responses and enhanced reliability across all endpoints.

    v1.1.0 ๐Ÿง 

    October 10, 2025

    ๐ŸŽ‰ Major Release: AI Enhancement Suite

    Vedika now learns from every query, predicts optimal configurations, and continuously monitors performance. Expected +40-50% improvement in efficiency, accuracy, and cost optimization over time.

    ๐ŸŒ 22 Languages Support

    • โ€ข 11 Indian Languages:
      Hindi, Bengali, Telugu, Tamil, Gujarati, Kannada, Malayalam, Marathi, Punjabi, Odia, Assamese
    • โ€ข 11 International Languages:
      English, Spanish, French, German, Italian, Portuguese, Russian, Japanese, Korean, Chinese, Arabic
    • โ€ข Context-Aware Translation:
      AI-powered translation preserves Sanskrit astrological terms across all languages
    • โ€ข Spiritual Accuracy:
      No generic machine translation - maintains cultural and spiritual context
    • โ€ข India Market Focus:
      Only astrology API with comprehensive Indian language coverage

    โญ Western Astrology Support

    • โ€ข Tropical Zodiac Birth Charts:
      Complete Western (season-based) astrology system
    • โ€ข 10 Planets Supported:
      Sun through Pluto (includes Uranus, Neptune not in Vedic)
    • โ€ข 13 House Systems:
      Placidus, Koch, Equal, Whole Sign, and 9 more professional systems
    • โ€ข Secondary Progressions:
      Long-term predictive technique (1 day = 1 year)
    • โ€ข Solar Returns:
      Annual birthday chart for yearly forecasts
    • โ€ข Synastry Analysis:
      Relationship compatibility with inter-chart aspects
    • โ€ข Current Transits:
      Compare natal positions with current planetary transits
    • โ€ข Independent Swiss Ephemeris:
      No dependency on external APIs (ยฑ0.01ยฐ precision)
    • โ€ข Competitive Advantage:
      Only API offering both Vedic AND Western astrology

    ๐Ÿ”Œ Real-Time WebSocket Features (Phase 4)

    • โ€ข Real-Time Transit Alerts:
      Subscribe to planetary transit notifications affecting your natal chart
    • โ€ข Streaming AI Predictions:
      Real-time streaming for astrological readings (powered by Vedika 2.1 Neural Engine)
    • โ€ข Live Credit Monitoring:
      Real-time credit balance updates as you use the API
    • โ€ข API Usage Tracking:
      Monitor your API calls in real-time with instant updates
    • โ€ข API Key Activity:
      Live monitoring of all your API keys' usage and statistics
    • โ€ข Socket.IO Integration: WebSocket server at ws://your-domain/ws with authentication
    • โ€ข Transit Alert Intelligence:
      Only notifies for transits within 2ยฐ orb (actively affecting you now)
    • โ€ข Severity Ratings:
      1-10 scale based on planet type, orb tightness, and aspect nature
    • โ€ข Auto-Updates:
      Transit calculations refresh every 6 hours automatically
    • โ€ข Multi-Framework Support:
      Examples for React, Vue.js, Next.js, and Python clients
    • โ€ข Complete Documentation:
      1,700+ lines of WebSocket API docs with examples
    • โ€ข Comprehensive Testing:
      19 integration tests covering all WebSocket features

    Phase 1: Persistent Memory System

    • โ€ข Self-Learning AI:
      Stores high-quality query patterns (โ‰ฅ7.0/10) in SQLite database
    • โ€ข Pattern Recognition:
      Learns from successful processing patterns and response quality
    • โ€ข Zero Performance Impact: Non-blocking storage (<1ms overhead)
    • โ€ข Expected Impact:
      +10-15% accuracy improvement over time

    Phase 2: Neural Pattern Predictor

    • โ€ข AI-Powered Predictions:
      Recommends optimal configurations before execution
    • โ€ข Smart Configuration:
      Predicts execution time, cost, and quality scores
    • โ€ข Confidence Scoring:
      Improves from 47% (10 patterns) to 90%+ (500+ patterns)
    • โ€ข Expected Impact:
      +15-25% efficiency improvement

    Phase 3: Performance Monitor

    • โ€ข Real-Time Tracking:
      Monitors execution time, quality, cost, and confidence
    • โ€ข Trend Analysis:
      Detects improving/declining/stable patterns
    • โ€ข Smart Recommendations:
      Generates optimization suggestions automatically
    • โ€ข Expected Impact:
      20-30% cost reduction through insights

    Safety & Reliability

    • โ€ข Feature flags for instant rollback (VEDIKA_ENABLE_*)
    • โ€ข Graceful degradation - errors never crash main workflow
    • โ€ข <1ms total overhead across all phases (<0.2% impact)
    • โ€ข Verified: Zero downtime, 100% test success rate

    Production Ready

    All phases deployed, tested, and operational. 10 patterns already collected with 8.5/10 average quality. Neural predictions active with 47% confidence (will improve to 90%+ as more patterns accumulate).

    v1.0.2

    October 10, 2025

    New Features

    • โ€ข Numerology Calculations:
      Added 35 Numerology calculations (Pythagorean + Chaldean systems)
    • โ€ข Homepage Visualization:
      Enhanced parallel processing visualization on homepage
    • โ€ข API Branding:
      Improved API branding consistency across all pages

    Improvements

    • โ€ข Cache Optimization:
      Optimized cache hit rate from 67% to 73% (17% cost savings)
    • โ€ข Response Time:
      Reduced P50 response time from 2.4s to 2.14s
    • โ€ข Dashboard UI:
      Enhanced dashboard UI with real-time usage metrics

    Bug Fixes

    • โ€ข Compatibility Analysis:
      Fixed compatibility analysis edge case for Nadi Dosha detection
    • โ€ข Timezone Conversion:
      Resolved timezone conversion issues for birth chart calculations

    v1.0.1

    October 3, 2025

    New Features

    • โ€ข Divisional Charts:
      Added support for 60 divisional charts (D1-D60)
    • โ€ข Dasha Systems:
      Implemented 7 Dasha systems with 5-level sub-periods
    • โ€ข MCP Integration:
      Added Model Context Protocol (MCP) for conversation persistence

    Improvements

    • โ€ข AI Upgrade:
      Upgraded to Vedika 2.1 Advanced AI for primary processing
    • โ€ข Yoga Detection:
      Enhanced Yoga detection accuracy to 96.4%
    • โ€ข Benchmarks Page:
      Added comprehensive benchmarks page with ROI calculator

    Security

    • โ€ข API Key Hashing:
      Implemented SHA-256 API key hashing
    • โ€ข PII Protection:
      Added PII masking in logs (GDPR compliance)
    • โ€ข Rate Limiting:
      Enhanced rate limiting: 500 requests per 15 minutes

    v1.0.0

    September 15, 2025

    Initial Release

    Initial Launch

    • โ€ข Parallel Processing:
      Multi-layer parallel processing system (82% better accuracy than GPT-4)
    • โ€ข Yoga Detection:
      300+ Yoga detection with classical validation
    • โ€ข Dosha Analysis:
      25 Dosha types with remedial suggestions
    • โ€ข Compatibility Matching:
      36-point Ashtakoota compatibility matching
    • โ€ข Astronomical Precision:
      NASA-grade astronomical precision (ยฑ0.001ยฐ)
    • โ€ข Multi-Region Deployment:
      Multi-region deployment (US-Central1, Asia-South1)
    • โ€ข Uptime SLA:
      99.97% uptime SLA with 8-minute MTTR
    • โ€ข REST API:
      Complete REST API with authentication

    Coming Soon

    • v1.3.0 (Q4 2025)

      Enhanced Dashboard with Advanced Analytics, Bulk Processing API

    • v1.4.0 (Q1 2026)

      Enterprise Team Management with Role-Based Access Control, White-Label API Customization

    • v2.0.0 (Q2 2026)

      Advanced Machine Learning for Pattern Recognition, Predictive Insights Engine

    Already Available: GraphQL API with Interactive Playground, Multi-language support (22 languages), Complete Western & Vedic Astrology, Muhurta & Panchang Calculations, Real-Time WebSocket Notifications, Token-Based Billing with Streaming, 300+ Yoga Detection, Complete Dosha Analysis, Dasha Systems, Compatibility Matching, Revenue Protection, Smart Query Routing, Webhook Support