Skip to content

See the new AI dev workflow first at NTL DEPLOY. Join live October 1st →

Pricing for AI features

Learn about pricing for AI features, such as Agent Runners and AI Gateway.

AI inference is a usage meter that measures the costs of using AI models and agents on Netlify. This usage meter calculates costs for using Agent Runners and AI Gateway.

AI inference is one of several usage meters, such as compute, bandwidth, and web requests, that convert usage to Netlify credits.

Note that the “Why did it fail?” troubleshooting failed deploys feature does not generate AI inference costs. If you use an Agent Runner to fix a failed deploy though, AI inference usage and costs will be calculated.

AI inference costs are calculated based on your AI model usage.

AI inference is measured based on the costs set by AI model providers, which convert AI model usage tokens to USD. Netlify then converts every $1 USD spent on AI model usage to 180 Netlify credits.

This rate is designed to make using AI features like Agent Runners and AI Gateway competitive with using AI providers without Netlify.

Feature/resourceUnitCredits ConsumedWhat this covers
AI inferenceAI model usage in USD (US Dollars)180 credits per $1 USDAgent Runners and AI Gateway

The costs of using Agent Runners is calculated using AI inference and compute as usage meters.

AI inference costs are determined by the AI model/s used in your agent runs. When you choose an agent for an agent run, the agent you choose picks an AI model based on the task type and complexity. To learn more about how AI models are priced, check out our AI model pricing rates table.

Agent Runners also use compute resources. Learn more about how compute is calculated in our how credits work docs.

The cost of using AI Gateway is calculated using only the AI inference usage meter.

When you use AI Gateway, you can choose a specific AI model to use for your requests.

For each request to the AI Gateway that ends successfully, the following is calculated:

  • Your account is billed for the amount of actual tokens used in that request.
  • The count of actual tokens used for each token type (input, output, cache reads, and cache writes) is converted to US dollar amounts, using the rates listed below.
  • The exact total dollar value for the request is converted to credits. Each dollar equals 180 credits.

There are no special credits for the AI Gateway. AI Gateway uses the same pool of credits included in your Netlify pricing plan, or any add-on credits you purchase.

To estimate your costs, you can check the Pricing per AI model table or explore our pricing calculator.

AI inference supports the following AI providers and models.

If you are using AI Gateway, you can choose a specific AI model to use for your requests.

If you are using Agent Runners, your AI agent will choose which AI models to use on your behalf to optimize performance and cost based on task type and complexity.

AI ProviderModel
OpenAIgpt-5
OpenAIgpt-5-codex
OpenAIgpt-5-mini
OpenAIgpt-5-nano
OpenAIgpt-4.1
OpenAIgpt-4.1-mini
OpenAIgpt-4.1-nano
OpenAIgpt-4o
OpenAIgpt-4o-mini
OpenAIo4-mini
OpenAIo3
OpenAIo3-mini
OpenAIcodex-mini-latest
Anthropicclaude-opus-4-1-20250805
Anthropicclaude-opus-4-20250514
Anthropicclaude-sonnet-4-5-20250929
Anthropicclaude-sonnet-4-20250514
Anthropicclaude-3-7-sonnet-20250219
Anthropicclaude-3-7-sonnet-latest
Anthropicclaude-3-5-haiku-20241022
Anthropicclaude-3-5-haiku-latest
Anthropicclaude-3-haiku-20240307
Googlegemini-2.5-pro
Googlegemini-flash-latest
Googlegemini-2.5-flash
Googlegemini-2.5-flash-preview-09-2025
Googlegemini-flash-lite-latest
Googlegemini-2.5-flash-lite
Googlegemini-2.5-flash-lite-preview-09-2025
Googlegemini-2.5-flash-image-preview
Googlegemini-2.0-flash
Googlegemini-2.0-flash-lite

You can also programatically access the up-to-date list in JSON format via a public API endpoint: https://api.netlify.com/api/v1/ai-gateway/providers.

The following table shows the current rates for AI tokens in USD (US dollars) for different AI providers before costs are converted to Netlify credits.

ProviderModelInput TokensCache readOutput TokensCache write
OpenAIgpt-5$1.25$0.125$10.00N/A
OpenAIgpt-5-codex$1.25$0.125$10.00N/A
OpenAIgpt-5-mini$0.25$0.025$2.00N/A
OpenAIgpt-5-nano$0.05$0.005$0.40N/A
OpenAIgpt-4.1$2.00$0.500$8.00N/A
OpenAIgpt-4.1-mini$0.40$0.100$1.60N/A
OpenAIgpt-4.1-nano$0.10$0.025$0.40N/A
OpenAIgpt-4o$2.50$1.250$10.00N/A
OpenAIgpt-4o-mini$0.15$0.075$0.60N/A
OpenAIo4-mini$1.10$0.275$4.40N/A
OpenAIo3$2.00$0.500$8.00N/A
OpenAIo3-mini$1.10$0.550$4.40N/A
OpenAIcodex-mini-latest$1.50$0.375$6.00N/A
Anthropicclaude-opus-4-1-20250805$15.00$1.500$75.00$18.75
Anthropicclaude-opus-4-20250514$15.00$1.500$75.00$18.75
Anthropicclaude-sonnet-4-5-20250929$3.00$0.300$15.00$3.75
Anthropicclaude-sonnet-4-20250514$3.00$0.300$15.00$3.75
Anthropicclaude-3-7-sonnet-20250219$3.00$0.300$15.00$3.75
Anthropicclaude-3-7-sonnet-latest$3.00$0.300$15.00$3.75
Anthropicclaude-3-5-haiku-20241022$0.80$0.080$4.00$1.00
Anthropicclaude-3-5-haiku-latest$0.80$0.080$4.00$1.00
Anthropicclaude-3-haiku-20240307$0.25$0.030$1.25$0.30
Googlegemini-2.5-pro$1.25$0.310$10.00N/A
Googlegemini-2.5-flash$0.30$0.075$2.50N/A
Googlegemini-2.5-flash-preview-09-2025$0.30$0.075$2.50N/A
Googlegemini-flash-latest$0.30$0.075$2.50N/A
Googlegemini-2.5-flash-lite$0.10$0.025$0.40N/A
Googlegemini-2.5-flash-lite-preview-09-2025$0.10$0.025$0.40N/A
Googlegemini-flash-lite-latest$0.10$0.025$0.40N/A
Googlegemini-2.5-flash-image-preview$0.30N/A$30.00N/A
Googlegemini-2.0-flash$0.10$0.025$0.40N/A
Googlegemini-2.0-flash-lite$0.075N/A$0.30N/A