> ## Documentation Index
> Fetch the complete documentation index at: https://docs.isaacus.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Prices

> Simple, transparent, and affordable legal AI

The Isaacus API has no upfront costs and no hidden fees. **You pay only for what you use.**

This page provides an overview of the pricing for each of our API endpoints. All amounts are in USD.

These prices may change in the future but we will always notify you before new pricing goes into effect.

These prices do not include any taxes that we may be required to collect by law in your particular jurisdiction — such taxes will be applied automatically to your invoices.

These prices apply only to the cloud-hosted Isaacus API. The pricing for our air-gapped Amazon SageMaker models is publicly available on [AWS Marketplace](https://aws.amazon.com/marketplace/seller-profile?id=seller-5e4iuidabgujc).

Those looking for on-premises deployments or a volume discount or to negotiate an alternative pricing model should [contact us](https://isaacus.com/contact/).

## Enrichment

Usage of our [enrichment models](/models#enrichment) is charged based on the number of tokens inputted into them.

Per token pricing for our enrichment models is as follows:

| Model                  | Price per token                                           | Description                                                 |
| ---------------------- | --------------------------------------------------------- | ----------------------------------------------------------- |
| **`kanon-2-enricher`** | $$\$3.50$$ per million tokens / $$\$0.0000035$$ per token | The first enrichment and hierarchical graphitization model. |

## Reranking

Usage of our [reranking models](/models#reranking) is charged based on the number of tokens inputted into them.

Per token pricing for our reranking models is as follows:

| Model                  | Price per token                                            | Description                                                                              |
| ---------------------- | ---------------------------------------------------------- | ---------------------------------------------------------------------------------------- |
| **`kanon-2-reranker`** | $$\$0.35$$ per million tokens / $$\$0.00000035$$ per token | The most accurate legal reranker on [Legal RAG Bench](https://arxiv.org/abs/2603.01710). |

## Embedding

Usage of our [embedders](/models#embedding) is charged solely based on the number of [tokens](/tokenization) inputted into them.

Per token pricing for our embedders is as follows:

| Model                  | Price per token                                            | Description                                                                                                          |
| ---------------------- | ---------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------- |
| **`kanon-2-embedder`** | $$\$0.35$$ per million tokens / $$\$0.00000035$$ per token | The most accurate legal embedding model on the [Massive Legal Embedding Benchmark (MLEB)](https://isaacus.com/mleb). |

## Extractive question answering

Usage of our [extractive question answering models](/models#extractive-question-answering) is charged based on the number of tokens inputted into them.

Per token pricing for our answer extractors is as follows:

| Model                        | Price per token                                           | Description                                                               |
| ---------------------------- | --------------------------------------------------------- | ------------------------------------------------------------------------- |
| **`kanon-answer-extractor`** | $$\$1.50$$ per million tokens / $$\$0.0000015$$ per token | Our base answer extractor, designed to balance precision with throughput. |

## Universal classification

Usage of our [universal classifiers](/models#universal-classification) is charged based on the number of tokens inputted into them.

Note that the number of tokens that get inputted into a universal classifier may differ from the number of tokens inputted into an API endpoint, for example, due to the addition of boilerplate tokens, as explained in our [costs page](/pricing/costs).

This table shows the price per token for each of our universal classifiers:

| Model                            | Price per token                                          | Description                                       |
| -------------------------------- | -------------------------------------------------------- | ------------------------------------------------- |
| **`kanon-universal-classifier`** | $$\$1.00$$ per million tokens / $$\$0.000001$$ per token | Our most powerful universal classification model. |

## On premises

If you're interested in running Isaacus models on your own infrastructure or in finetuning them on your own data, feel free to [reach out](https://isaacus.com/contact/) to learn more about our on-premises solutions.