Compute Points

Overview

Compute Points are a measure of the computational resources used when making queries in Bind AI. Each subscription tier is allotted a fixed number of compute points, which refresh every month. You can view the compute points utilized per model here.

How Compute Points Work

Compute points are consumed based on several factors:

• The model you select.

• The length of your query (number of tokens in the query).

• The length of the response to your query (number of tokens in the response).

• The length of the history or context used (e.g., previous messages or embeddings retrieved).

Why Compute Points?

Bind AI supports over 15 AI models from multiple providers, each with different compute requirements. Compute points allow you the flexibility to choose the model of your choice based on how many compute points you have available. More computationally intensive models consume more compute points.

Example

For a query like "Hello how are you," which consists of approximately 4 tokens:

• Using Claude 3.5 Sonnet might consume ~8 compute points.

• Using GPT-4o mini might consume 4.8 compute points.

Plan Limits

Each subscription tier has a different allocation of compute points. For example, the Lite tier may have 200K compute points, while the Premium tier has 1M compute points.

Important Notes

• Compute points are not shown in real-time, but you will receive a notification when you consume your points.

• The compute points per token may vary based on model availability and efficiency.

Compute Points vs Total Tokens Ingested

• Total Tokens Ingested measures the amount of data synced and stored for context, while Compute Points measure the computational resources used for processing queries.

• Total Tokens Ingested is related to data integration, whereas Compute Points are related to query execution and model usage.

• Plan limits for Total Tokens Ingested and Compute Points are separate and should be managed according to your subscription tier.

How do I get unlimited compute points?

You can add your own API keys to get unlimited queries. You can find this option in the navigation where it says “Add API keys”. Currently, API keys are supported only for Anthropic (Claude) and OpenAI models. You can also maximize your usage by picking the models which require less compute points and still have superior reasoning capabilities (e.g. Mistral Codestral is a very competent model for coding, and requires almost half the compute points as compared to Claude 3.5 Sonnet)

For further assistance or questions about Total Tokens Ingested or Compute Points, please contact Bind AI support.