Flexible pricing to let you start and grow

Choose the plan that best fits your needs.

Free

Perfect for developers who want to try out our platform and experiment with AI models.

Free up to $10 per month

  • Access to 50+ models
  • Experiment with AI
  • Community support
Create account

Utility

Ideal for businesses looking to scale their AI operations with flexible usage-based pricing.

Starts from $10 per month

  • Priority inference and rate limits
  • Pay for usage via auto charges
  • Email support
Create account

Enterprise

For large organizations requiring custom solutions and enterprise-grade reliability.

Custom

  • Custom payment options
  • Dedicated AI Engineering support
  • SLA guarantees
Contact us

Utility prices

FeatureDescriptionBenefitsPrice (excl. VAT)
InferenceInteract with +75 models with one API key

  • API key management
  • Fallback models and retries
  • Structured output guarantees
  • OpenAI compatible API
  • Interoperable, structured AI call
  • Centralized billing
  • Custom model management
Current model pricing

+ 10% platform fee

TracesComprehensive AI observability
  • Insight into prompts, inputs and outputs
  • Performance, latency and error monitoring
  • Metadata support
  • 30 days retention
$0.50 per 1k spans
IndexesVector indexing and search
  • Custom knowledge integration
  • Vector search capabilities
  • Automatic chunking

Current embedding pricing

+ 30% platform fee

EvaluationsIntegrated evaluation and testing suite
  • Automatic LLM as a judge
  • Custom evaluation metrics
  • Model testing and comparison
  • Quality assurance automation
Free when Beta
AnalyticsUsage analytics and cost tracking
  • Cost tracking and analysis
  • Usage auditing per model
  • Project-level analytics
  • End-user usage tracking
  • Cost allocation and reporting
Free when Beta

Frequently Asked Questions

What is included in the free tier?

The free tier includes up to $10 worth of usage per month, which covers all services including inference, traces, indexing, and context retrieval. This is perfect for developers who want to try out the platform or build small projects.

How does the inference pricing work?

Inference costs are based on the model you choose and include a 10% proxy fee. The proxy provides enhanced capabilities like fallback models, retries, and structured output guarantees. You can find current model pricing in our documentation.

What are traces and how are they priced?

Traces provide comprehensive AI observability, including insight into prompts, inputs, outputs, and performance monitoring. They are priced at $0.50 per 1,000 spans with 30 days retention.

How are indexing and context retrieval priced?

Indexing and context retrieval use our embedding model, priced at $0.13 per 1M tokens plus a 30% platform fee. Indexing converts documents at 2 tokens per byte, while context retrieval is based on the number of tokens retrieved.

What features are currently free?

Currently, Evaluations and Analytics are free. Evaluations include automatic LLM judging, custom metrics, and model testing. Analytics provides comprehensive usage tracking, cost analysis, and project-level insights.

How do charges work?

For the utility tier, you can configure automatic top-ups of your Opper credit balance or issue one-time top-ups. Automatic top-ups ensure uninterrupted service while keeping payments manageable. There's a minimum charge of $10 per month for the paid tier.

How secure are my payments?

All transactions are handled by our payment processor Stripe. Stripe ensures you get industry-standard receipts, PCI compliance, and enterprise-grade payment security. Your payment information is never stored on our servers.

Where is Opper hosted?

The Opper platform is hosted in AWS's Stockholm (EU) region with enterprise-grade security. This ensures compliance with GDPR and other EU data protection regulations.