Intro to Dust
What is Dust and how to use the platform
What is Dust?
Dust is a platform that enables teams to create customizable and secure AI agents powered by leading large language models. These agents integrate with company data sources to enhance productivity and improve work processes across various domains
Ask your first question
To use an agent, simply type your message and mention the agent with "@".
Ask @dust to find information within your company knowledge base
Start asking questions to @dust. This agent is setup by default and will use all of the internal documentation integrated to Dust to generate the answer.
![](https://files.readme.io/c3ff6fe796f45878ac1e82d24e265c7359336aaa390897fd4f0b8fa43a492b45-Screenshot_2024-08-29_at_12.07.42.png)
Talk to the best LLMs
Dust acts as replacement for ChatGPT, Gemini and other similar tools. Simply call one of the generic agents (ex: @gpt) to use it.
![](https://files.readme.io/8a2081f51eeba255a964aaeb9906d2340411659f68b55e385c13eceddb870e09-Screenshot_2024-08-29_at_12.23.09.png)
➡️ Best 500+ ways to use @GPT, @Claude & @Gemini
Talk to specialized agents built by you or your colleagues
You can now browse the other agents your team already created or start creating your own.
Lean more on any agents (instructions, tools, usage) by clicking on the 3 dots.
![](https://files.readme.io/86b031fb4b3e3dd3e2d9e60d1639802c9f9383c0e8bcc8c3a874d3df66132acb-Screenshot_2024-08-28_at_16.02.06.png)
Talk to multiple agents in the same conversation
Dust allows you to call multiple agents in one discussion and have them work together to achieve a given task.
Example 1: Ask two agents to perform the same task
Extremely helpful to benchmark models and agents against each other.
![](https://files.readme.io/8db5b6c11123eb75066d1976b2e4088f173fafdbae00f98947cbe7d24d0896ce-Screenshot_2024-08-29_at_12.19.05.png)
Example 2: Use agents sequentially to build on each other's work.
Extremely helpful when you have a set of specialized tasks to perform (e.g. ask an @accountSummary agent to generate a summary on an account, then ask @emailFollowup to draft an email followup. (Learn more in the use cases tab)️
![](https://files.readme.io/0f407e3180680c338ad9934772e819514e79a3eea8e213b119d63b9c4b8e51a6-Screenshot_2024-08-29_at_12.14.50.png)
Updated 3 days ago
Now, you probably want to know how to build your first agent. Discover our use cases & tutorials to make it happen.