Pricing
Starter
For you and your projects
99 a month7 days free, no card needed
Worth up to 150 at pay-as-you-go prices
Start free trialIncludes:
- Saudi-hosted models
- Global models (Gemini, Claude, GPT)
- alOCR, alRetrieval and alGuardrail
- Playground
- Dashboard and request logs
- One request at a time
- 60 requests per minute (Rate limit)
- Email support
Pro
For startups building on AI
1,499 a month
Worth up to 3,000 at pay-as-you-go prices
Get ProEverything in Starter, plus:
- 20x more usage than Starter*
- 5 requests at once (Concurrency)
- 300 requests per minute (Rate limit)
- Up to 5 team members
- Priority support
Enterprise
For large organizations and companies
Custom
Everything in Pro, plus:
- Custom usage and limits
- Unlimited team members
- Dedicated, with a service agreement (SLA)
- Run on your own servers (On-premise)
*Usage and rate limits apply. See how usage works below
*Prices exclude 15% VAT. Billed monthly, cancel any time
*Prices, plans and models are subject to change
What a month of usage covers
Choose a model or tool to see how much a full month of usage covers if you spend it all there
Token amounts assume 3 input tokens for each output token.
Pay only for what you use. Every model and tool has its own price, per 1M tokens unless stated
Models
| Model | Hosting | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
 deep-sa/alEmbeddingSaudi-hostedIn $0.05 |
Saudi-hosted | $0.05 | — |
 llama-3.3-70bSaudi-hostedIn $0.15 · Out $0.48 |
Saudi-hosted | $0.15 | $0.48 |
 gpt-oss-120bSaudi-hostedIn $0.22 · Out $0.90 |
Saudi-hosted | $0.22 | $0.90 |
| gpt-oss-20bSaudi-hostedIn $0.11 · Out $0.45 |
Saudi-hosted | $0.11 | $0.45 |
 qwen/qwen3.8-27bSaudi-hostedIn $0.38 · Out $3.83 |
Saudi-hosted | $0.38 | $3.83 |
| deep-sa/alEmbedding-2.0Saudi-hostedIn $0.05 |
Saudi-hosted | $0.05 | — |
 minimaxai/minimax-m2.7Saudi-hostedIn $0.31 · Out $1.26 |
Saudi-hosted | $0.31 | $1.26 |
| openai/gpt-6-lunaGlobalIn $0.13 · Out $0.62 |
Global | $0.13 | $0.62 |
 google/gemini-3.8-flashGlobalIn $0.94 · Out $4.69 |
Global | $0.94 | $4.69 |
| google/gemini-3.1-proGlobalIn $2.50 · Out $15.00 |
Global | $2.50 | $15.00 |
 anthropic/claude-sonnet-5.5GlobalIn $2.50 · Out $12.50 |
Global | $2.50 | $12.50 |
 z-ai/glm-5.3-flashGlobalIn $0.19 · Out $0.62 |
Global | $0.19 | $0.62 |
Tools
| Tool | Hosting | Price |
|---|---|---|
| alGuardrailSaudi-hostedFree for now |
Saudi-hosted | Free for now |
| alOCRSaudi-hosted$6.67 per 1,000 pages |
Saudi-hosted | $6.67 per 1,000 pages |
| alRetrievalSaudi-hostedBilled as the alOCR, alEmbedding and model use it makes |
Saudi-hosted | Billed as the alOCR, alEmbedding and model use it makes |
*Rate limits apply
*Prices shown exclude 15% VAT
*Prices, plans and models are subject to change
Frequently Asked Questions
- How is usage counted?
- Each call is counted by its tokens at the model's price, or by pages for alOCR. It comes out of your plan's usage, or from your wallet at pay-as-you-go prices if you have no plan or your plan's usage is used up. alGuardrail is free
- Do different models have different costs?
- Yes. Each model has its own price per 1M input and output tokens, listed under Pay-as-you-go. The same usage goes further on smaller models
- How do usage limits work?
- Each plan has a monthly usage that renews every month. When it is used up, calls go on from your wallet at pay-as-you-go prices, and stop if the wallet is empty until the usage renews. Plans also limit requests per minute and requests at once
- How is "worth up to" worked out?
- It is what a month of usage would cost at pay-as-you-go prices if you spent it on Saudi-hosted models and our tools
- Can I get a refund?
- For refund requests or billing questions, please contact us at alapi@deep.sa