API Pricing
Transparent pay-as-you-go pricing for Atlas Cloud AI models
Pricing Model
Atlas Cloud uses transparent pay-as-you-go pricing. No minimum commitments, no hidden fees.
Billing Rules
| Model Type | Billing Unit | How It Works |
|---|---|---|
| LLM (Chat) | Per million tokens | Based on input + output token count |
| Image Generation | Per generation | Varies by model, resolution, and parameters |
| Video Generation | Per generation | Varies by model, duration, and resolution; some models are billed by output video tokens |
| Audio Generation | Per generation | Varies by model |
| Image Tools | Per operation | Background removal, upscaling, face swap, etc. |
- LLM usage is charged after each request, based on the actual token usage in the response
- Image/Video generation is charged per request; failed tasks are automatically not charged
For a detailed explanation of how each model type is charged — including cached-token pricing, token-billed video models, and the cost estimation API — see How Models Are Billed.
Model Prices
View real-time pricing for all 400+ models:
You can also view pricing on each model's detail page in the Model Library.
Volume Discounts
We offer competitive rates for high-volume customers.
Estimate your usage
Calculate your expected monthly API consumption across all models.
Contact sales
Email [email protected] with your estimated volume.
Get custom pricing
Receive a tailored pricing plan based on your usage patterns.
Developer Plans
Temporarily Unavailable
Developer subscription plans are currently offline. Stay tuned for official announcements on when they will return. For high-volume needs, contact [email protected] for a custom plan.
Cost Optimization Tips
- Choose the right model: More expensive doesn't always mean better for your use case. Test multiple models to find the best quality-to-cost ratio
- Use smaller models for simple tasks: For basic Q&A or classification, a smaller, cheaper LLM may be sufficient
- Set
max_tokens: Limit response length for LLM calls to avoid unnecessary token usage - Batch efficiently: Group similar requests together and process them efficiently
- Monitor usage: Track your spending in the Console to identify optimization opportunities
Payment
- Top up at Billing via credit card, WeChat Pay, Alipay, or cryptocurrency
- Minimum top-up: $25
- Auto top-up available to prevent service interruptions
- Credits expire 365 days from purchase
For full billing details, see Pricing & Billing.