Billing & Plans
- How does pay-as-you-go pricing work for LLM APIs?
You pay per token used, with no upfront commitment. Input and output tokens are priced separately, and you're billed monthly based on your usage. Rates vary by model and provider. - What is the best way to monitor LLM API usage and costs?
Use your provider's built-in usage dashboard and set up automated alerts. For more control, route requests through a proxy like Helicone or OpenMeter to track costs per model, user, or feature. - Can I set spending limits on my LLM API account?
Yes, most providers let you set hard or soft spending limits. OpenAI, for example, has monthly budget caps, while others offer similar controls or require you to set them via a proxy. - Why did my LLM API bill spike unexpectedly?
Common causes include a bug causing repeated calls, increased traffic, a switch to a more expensive model, or a change in usage patterns. Check your usage logs for anomalies first. - Is it cheaper to prepay for LLM API credits?
Sometimes. Some providers offer discounts for prepaying, but many don't. Compare the discount to the opportunity cost of tying up cash and the risk of unused credits. - How do I get a refund for unused LLM API credits?
Most providers do not offer refunds for unused credits, but some may make exceptions. Check your provider's policy and contact support with a clear explanation.