Simple, Transparent Pricing
Pay only for what you use. No hidden fees.
Credit Calculator
Minimum purchase: 50K credits for USD 5
Purchased credits are valid for one year from the purchase date.
Base Credits
0
Bonus Credits
0
Total Credits
0
How credits are used
Each AI answer uses credits based on its AI model category and the tokens it reads and writes.
Managed AI
Qlar manages model access. Model usage is paid from your Qlar credits.
Connected AI
Connect your own provider account (API key or subscription). The provider bills you directly for the model; Qlar credits cover only orchestration and application usage.
AI response rates
Credits per 1,000 tokens
| Category | AI Model | Managed AI* | Connected AI | ||
|---|---|---|---|---|---|
| Input / output | Minimum per answer | Input + output | Minimum per answer | ||
| Lightweight | Light, Smart | 4 / 20 | 15 | 3 | 10 |
| Balanced | Analyst, Insight, Thinker | 50 / 250 | 175 | 60 | 100 |
| Powerful | Sage | 300 / 1,500 | 1,000 | 300 | 500 |
- * Managed AI figures are estimates for uncached input, not fixed token tariffs.
- Each answer is charged the greater of its usage and the category minimum, never both added together.
- Tokens are measured proportionally, not in rounded blocks of 1,000. Connected AI uses one rate for input and output tokens combined, and cached input reads are not counted.
- Tools, storage and connections are charged separately.
- Knowledge Search API — 1 credit per query (free with your own embedding model).
Example calculation
Balanced category with Managed AI: 50 credits per 1,000 input tokens, 250 per 1,000 output tokens, minimum 175.
Usage = (input tokens ÷ 1,000 × input rate) + (output tokens ÷ 1,000 × output rate)
Charge = the greater of usage or the minimum
| Answer | Input tokens | Output tokens | Usage | Minimum | Charged |
|---|---|---|---|---|---|
| Typical answer | 3,000 | 500 | 3 × 50 + 0.5 × 250 = 275 | 175 | 275 credits ($0.0275) |
| Short answer | 1,000 | 200 | 1 × 50 + 0.2 × 250 = 100 | 175 | 175 credits ($0.0175) |
All model calls within one answer (planning, query generation, answer writing) are added up before the minimum is applied.
Need an enterprise plan?
Contact us to discuss volume, deployment and terms for your organization.