Memory Index is publicly available

Date: June 19, 2024

Content

High Dimensional Research announced public access to Memory Index today. The system functions as "a repository of user-anonymised web trajectories -- storing page structures as a graph, and navigation across those structures as searchable data."

How It Works

When AI models use Nolita to complete web-based tasks, they can "share and recall patterns of navigation from other successful objectives when accomplishing subsequent tasks."

What to Expect

As the Memory Index expands, performance improves incrementally. Early improvements manifest subtly, then refine toward more sophisticated step application over time. The team continues enhancing search and replay functionality across both the Index and Nolita platform.

Getting Started

Users can register at the HDR dashboard. Default allocation includes 10,000 monthly requests, with additional capacity available through subscription.

Setup process:

• Execute npx nolita auth
• Generate an API key via the dashboard
• Input credentials into Nolita (saved locally)

For package imports:

const agent = makeAgent(
  { apiKey: process.env.OPENAI_API_KEY!, provider: "openai" },
  { model: "gpt-4" }
);

const browser = await Browser.launch(false, agent, undefined, {
  apiKey: process.env.HDR_API_KEY,
});

const page = await browser.newPage();

Upcoming Features

• nolita record and nolita replay commands for applying navigation trajectories across sessions
• Dashboard memory management tools for organizations
• Agent engine rewrite (pre-alpha) with Claude 3 support planned
• Local model integration anticipated following the engine rewrite