SkyAIAppPricing & Usage Billing
Transparent pricing, pay as you grow
Pricing Philosophy
SkyAIApp pricing is built on 'fair transparency and pay-as-you-go' principles. We benchmark against industry leaders (Portkey, LangSmith, Alibaba Cloud) and offer 2-6% lower rates. All costs are visible in real-time on your dashboard with no hidden fees.
1. Base Routing Fees
Base fee for each API request routed to model providers. Tiered pricing with volume discounts.
| Request Volume | Price (per million) | vs Industry |
|---|---|---|
| 0 - 1M requests | $1.50 | -4% (industry: $1.56) |
| 1M - 10M requests | $1.20 | -6% (industry: $1.28) |
| 10M+ requests | $1.00 | -5% (industry: $1.05) |
2. Semantic Cache Fees
Semantic cache dramatically reduces model call costs. Cache hits require only minimal read fees instead of full model calls.
# 缓存节省示例计算 / Cache savings example
original_cost = 10000 * 0.003 # 10k requests × $0.003/call = $30
cache_hit_rate = 0.6 # 60% 缓存命中率
cache_hits = 10000 * 0.6 # 6000次命中
cache_misses = 10000 * 0.4 # 4000次未命中
# 使用缓存后的成本 / Cost with cache
cache_read_cost = cache_hits * 0.00001 # $0.01/M tokens ≈ $0.01/100 calls
cache_miss_cost = cache_misses * 0.003
total_with_cache = cache_read_cost + cache_miss_cost # ≈ $12.06
savings = original_cost - total_with_cache # ≈ $17.94 (59.8%)3. Tracing & Logging Fees
Observability is critical in production. We offer tiered tracing and logging services to match your needs.
Basic Tracing
Recommended- Request/response metadata
- Latency, cost, tokens
- 7-day retention
Detailed Log Storage
- Full request/response content
- Custom retention
- Real-time streaming
4. Agent Runtime Fees
Running AI Agents requires compute and memory. We offer elastic, per-second billing.
Compute Time
Additional Memory
5. Evaluation Services
Use our evaluation framework to validate model output quality, accuracy, and performance.
📊 Complete Billing Example
Assume your AI application has the following monthly usage:
💳 Payment & Billing
Billing Cycle
Monthly billing, invoices generated on the 1st. Accepts credit cards, ACH, wire transfer. Enterprise customers can request invoicing (subject to approval).
Real-time Dashboard
Dashboard shows real-time usage and estimated costs for the current billing period. View cost breakdown by project, environment, and user.
Budget Alerts
Set monthly budget caps, receive email/webhook alerts at 80%/90%/100%. Configure overage behavior (continue or pause service).
❓ FAQ
How are model API costs calculated?
Model API costs are charged directly by model providers (OpenAI, Anthropic, etc.) at their official rates. SkyAIApp does not mark up model costs—we only charge for platform routing and value-added services listed above.
How do I view detailed billing?
Login to Console → Billing → Usage Details to view hourly/daily breakdowns. Filter by project, environment, model, or user. Export as CSV.
Is there a free tier?
Yes! The Free plan includes up to $20 equivalent in monthly platform usage (~13M base routing requests). Overages billed at rates above. Model API costs are separate.
Do enterprise customers get discounts?
Yes. Annual enterprise customers ($20k+) receive 10-15% discounts, plus dedicated support, SLA guarantees, and custom features. Contact hello@skyaiapp.com.
Start Now, Pay As You Grow
Start free with no credit card required. Free plan includes $20/month platform usage. Upgrade to Pro or Team anytime, pay only for what you use.
Was this page helpful?
Let us know how we can improve
Was this page helpful?
Let us know how we can improve