Knowledge Base
GPTWeb Cost Structure
GPTWeb uses a unique two-part cost structure that gives you complete transparency and control. Your total monthly cost comes from the GPTWeb platform subscription PLUS your direct payments to service providers through our BYOK (Bring Your Own Keys) model.
How Your Monthly Cost Breaks Down
Part 1: GPTWeb Platform (Paid to GPTWeb)
• Base subscription tier
• Included tokens for platform usage
• Token overages if you exceed your plan
• Visitor overages if you exceed limits Part 2: BYOK Provider Costs (Paid Directly to Providers)
• AI/LLM costs (Anthropic Claude, OpenAI)
• Email service (SendGrid, Mailgun, Amazon SES)
• Enrichment services (People Data Labs, Clearbit, ZoomInfo - optional)
• SMS services (Twilio - optional)
• Base subscription tier
• Included tokens for platform usage
• Token overages if you exceed your plan
• Visitor overages if you exceed limits Part 2: BYOK Provider Costs (Paid Directly to Providers)
• AI/LLM costs (Anthropic Claude, OpenAI)
• Email service (SendGrid, Mailgun, Amazon SES)
• Enrichment services (People Data Labs, Clearbit, ZoomInfo - optional)
• SMS services (Twilio - optional)
GPTWeb Platform Pricing
| Plan | Monthly | Annual | Annual Savings | Tokens/Month | Visitors/Month | Custom Domains | RAG Docs | Crawler Pages |
|---|---|---|---|---|---|---|---|---|
| Free Trial | $0 | $0 | N/A | 10,000 | 1,000 | 0 | 10 | 100 |
| Starter | $199.95 | $2,399 | ~$400/year | 2,000,000 | 1,000 | 1 | 100 | 500 |
| Pro ⭐ | $599 | $6,999 | ~$1,200/year | 5,000,000 | 1,000 | 3 | 200 | 1,000 |
| Enterprise | $2,995 | $34,995 | ~$1,000/year | 10,000,000 | 1,000 | 5 | 300 | 2,000 |
Overage Costs
| Plan | Token Overage | Visitor Overage | Support Level |
|---|---|---|---|
| Free Trial | Upgrade Required | Upgrade Required | Community |
| Starter | $0.15 per 1K tokens | $0.25 per visitor | |
| Pro ⭐ | $0.015 per 1K tokens | $0.25 per visitor | Priority |
| Enterprise | $0.015 per 1K tokens | $0.20 per visitor | Dedicated Manager |
BYOK Provider Costs (Paid Directly)
| Category | Provider Options | What It Powers | You Pay |
|---|---|---|---|
| AI/LLM | Anthropic, OpenAI | Claude conversations, embeddings for semantic search | Provider directly |
| SendGrid, Mailgun, Amazon SES | Agent workflows, DQL alerts, email campaigns | Provider directly | |
| Enrichment | People Data Labs, Clearbit, ZoomInfo | Contact and company enrichment (optional) | Provider directly |
| SMS | Twilio | SMS notifications (optional) | Provider directly |
10x Lower
Pro Plan Token Savings
40-70%
Cache Hit Rate Potential
14 Days
Free Trial Duration
Why BYOK Saves You Money
💰 Full cost visibility - See exactly what AI and services cost
🔓 No vendor lock-in - Switch providers anytime
⚡ Latest models - Access new AI models immediately
📊 No markup - We don't profit from your AI or email usage Unlike competitors who charge per-seat or hide AI costs in their pricing, you control your spending directly and can optimize based on your actual usage.
🔓 No vendor lock-in - Switch providers anytime
⚡ Latest models - Access new AI models immediately
📊 No markup - We don't profit from your AI or email usage Unlike competitors who charge per-seat or hide AI costs in their pricing, you control your spending directly and can optimize based on your actual usage.
Cost Optimization Tips
1. Enable Response Caching
GPTWeb caches common questions to reuse answers. First visitor asks about pricing, AI generates it. Next 100 visitors get the cached answer with no token cost. This typically reduces token usage by 40-70%. 2. Choose the Right AI Model
• Claude Haiku - Simple Q&A, high volume ($)
• Claude Sonnet - Most use cases, best balance ($$)
• Claude Opus - Complex reasoning ($$$$) 3. Curate Your Knowledge Base
Smaller, focused knowledge bases mean faster searches and fewer tokens per query. Upload targeted documents, not entire websites. 4. Use the Crawler Wisely
Set depth limits and exclude patterns to avoid crawling unnecessary pages.
GPTWeb caches common questions to reuse answers. First visitor asks about pricing, AI generates it. Next 100 visitors get the cached answer with no token cost. This typically reduces token usage by 40-70%. 2. Choose the Right AI Model
• Claude Haiku - Simple Q&A, high volume ($)
• Claude Sonnet - Most use cases, best balance ($$)
• Claude Opus - Complex reasoning ($$$$) 3. Curate Your Knowledge Base
Smaller, focused knowledge bases mean faster searches and fewer tokens per query. Upload targeted documents, not entire websites. 4. Use the Crawler Wisely
Set depth limits and exclude patterns to avoid crawling unnecessary pages.
Common Questions
Do I pay GPTWeb for AI usage?
Yes and No. GPTWeb tracks tokens for fair platform usage billing, but you ALSO pay AI providers directly using your own API keys. This gives you complete visibility into actual AI costs. What if I have a traffic spike?
Contact us! We can adjust your limits mid-cycle. We'd rather help you succeed than surprise you with overages. Can I change plans anytime?
Yes. Upgrade instantly, downgrade at the end of your billing cycle. Is there a free trial?
Yes! 14-day free trial with full features. No credit card required.
Yes and No. GPTWeb tracks tokens for fair platform usage billing, but you ALSO pay AI providers directly using your own API keys. This gives you complete visibility into actual AI costs. What if I have a traffic spike?
Contact us! We can adjust your limits mid-cycle. We'd rather help you succeed than surprise you with overages. Can I change plans anytime?
Yes. Upgrade instantly, downgrade at the end of your billing cycle. Is there a free trial?
Yes! 14-day free trial with full features. No credit card required.
Ready to see your actual costs? Start a free trial or contact sales@gptweb.com for custom enterprise pricing with volume discounts and dedicated support.
GPTWeb is the future of engagement, websites and marketing automation combined that is built for the AI era, built for now.