Simple, transparent pricing.
Start at $19. Not $100. Not $700.
Token-based consumption — pay for what you use, at a fraction of the competition.
Free
Developer sandbox — explore the platform at no cost
- 1.25M tokens/mo
- 25 pages ingested
- 0.01 GB storage
- 1 seat
Personal
For individuals and freelancers
- 10M tokens/mo
- 200 pages ingested
- 0.5 GB storage
- 1 seat
- Google Web Search add-on available
Starter
For small teams and early-stage startups
- 25M tokens/mo
- 1,000 pages ingested
- 2 GB storage
- 5 seats
- Google Web Search add-on available
Business
For small businesses
- 75M tokens/mo
- 4,000 pages ingested
- 5 GB storage
- 20 seats
- Google Web Search add-on available
Growth
For scaling businesses
- 200M tokens/mo
- 10,000 pages ingested
- 20 GB storage
- 50 seats
- Google Web Search add-on available
Enterprise
Single-tenant. Custom SLAs. Advanced LLM models.
- Tailored token allocation
- Custom pages & storage
- SSO / SAML, HIPAA BAA
- Dedicated CSM & SLA guarantee
Transparent token-based consumption
Like Progress Agentic RAG, we meter actual usage beyond your plan's token budget. Unlike them — we charge $0.005 per 1,000 tokens, not $8.00 per 1,000. Our tokens are real LLM tokens you can verify directly against your provider's bill.
| Resource | Our rate | Progress Agentic RAG | Saving |
|---|---|---|---|
| Token consumption | $0.005 / 1k tokens | $8.00 / 1k tokens | 99.99% |
| Page ingestion (fast) | $0.005 / page | N/A | — |
| Page ingestion (hi-res) | $0.015 / page | N/A | — |
| Storage | $0.150 / GB/mo | N/A | — |
| Google Web Search | $15.00 / 1k searches | N/A | — |
Google Web Search add-on
Enable per Knowledge Base on any paid plan. Each RAG query fires 1 live Google search and injects the results into the LLM context — grounding your answers in current, real-world information.
Billed at $15.00 per 1,000 searches as a separate line item on your monthly invoice. No minimum. Disable anytime.
📄 How pages are calculated
1 page = 3,000 characters. Fractional pages are fully supported. A 1,500-character document counts as 0.5 pages. PDFs and PowerPoint files use their native page/slide count. All other file types divide character count by 3,000.
The same power. A fraction of the price.
| Competitor | Their price | Our equivalent | You save |
|---|---|---|---|
| Ragie.ai Starter | $100/mo | Personal ($19/mo) | 81% |
| Ragie.ai Pro | $500/mo | Business ($149/mo) | 70% |
| Progress Agentic RAG Starter | $700/mo | Business ($149/mo) | 79% |
| Progress Agentic RAG Pro | $1,925/mo | Growth ($399/mo) | 79% |
Full feature comparison
| Feature | Free | Personal | Starter | Business | Growth | Enterprise |
|---|---|---|---|---|---|---|
| Retrieval | ||||||
| Hybrid Search (BM25 + vector) | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Reranking | — | ✓ | ✓ | ✓ | ✓ | ✓ |
| Hierarchical Search | — | — | ✓ | ✓ | ✓ | ✓ |
| Recency Bias | — | — | ✓ | ✓ | ✓ | ✓ |
| Documents | ||||||
| Hi-Res Ingestion | — | — | — | ✓ | ✓ | ✓ |
| Entity Extraction | — | ✓ | ✓ | ✓ | ✓ | ✓ |
| Multi-Language | — | ✓ | ✓ | ✓ | ✓ | ✓ |
| Platform | ||||||
| Multi-Tenant Partitions | — | — | — | ✓ | ✓ | ✓ |
| API Connectors | — | — | — | ✓ | ✓ | ✓ |
| Webhooks | — | — | — | ✓ | ✓ | ✓ |
| Google Web Search | — | ✓ | ✓ | ✓ | ✓ | ✓ |
| Analytics & Admin | ||||||
| Basic Analytics | — | — | ✓ | ✓ | ✓ | ✓ |
| Advanced Analytics | — | — | — | ✓ | ✓ | ✓ |
| White-Label / Custom Domain | — | — | — | — | ✓ | ✓ |
| Custom Prompt Templates | — | — | — | — | ✓ | ✓ |
| SSO / SAML | — | — | — | — | — | ✓ |
| Uptime SLA | — | — | — | — | ✓ | ✓ |
Questions & answers
What counts as a token?
A token is a unit of LLM processing — roughly 4 characters of English text. Every RAG query consumes tokens for the system prompt, retrieved context, your question, and the AI's answer across all pipeline agents. On average, a full query uses about 40,000 tokens. Your plan's token budget covers this full cost. You can see your exact consumption in your billing dashboard.
How are pages calculated for document ingestion?
1 page = 3,000 characters. Fractional pages are supported — a 1,500-character document counts as 0.5 pages. PDFs and PowerPoint files use their native page or slide count. All other formats (Word, Markdown, CSV, HTML, etc.) divide total character count by 3,000.
What happens if I exceed my plan limits?
On the Free plan, usage is hard-capped — once your token budget is spent, queries return an error until the next billing period. On paid plans, usage beyond your included budget is billed at the consumption rates shown above. You can see a real-time forecast on your billing dashboard and set a spending cap to prevent surprise charges.
How does Google Web Search billing work?
Web Search is opt-in per Knowledge Base. When enabled, each RAG query fires 1 Google search to pull live results into the LLM context. You're charged $15.00 per 1,000 searches consumed — each web-enabled query = 1 search = $0.015. This appears as a separate line item on your invoice. No minimum commitment, and you can disable it anytime per Knowledge Base.
Can I switch plans at any time?
Yes. Upgrades take effect immediately with prorated billing. Downgrades take effect at the end of your current billing period so you keep your current limits through to renewal.
Is there a free trial?
The Free plan is permanently free — no credit card required, no time limit. It includes 1,250,000 tokens (~31 full queries) and 25 pages of document ingestion. When you're ready to scale, upgrade to Personal for $19/month.
What is hi-res ingestion?
Hi-res ingestion uses a specialized vision model to extract text from scanned PDFs, complex layouts, diagrams, and images embedded in documents. It costs more than fast ingestion but produces significantly better results for image-heavy content. Available on Business and Growth plans.