Dust's latest capability, structured data analysis with table queries, marks a significant leap forward to help anyone in the company make the most of data.

In this article, I'll expose the core motivations behind building Dust, how we believe the generative AI ecosystem will evolve and why we are building products at the intersection of models, human and productivity (instead of training models despite all the investor market pull to do so). There

2023 was a year of exciting exploration for many teams as they sought to understand the capabilities of generative AI (“GenAI”) and the impact it could have on what they sell but also on how they work to sell it. In 2024, we expect many companies to decide on a

The convergence of models' interfaces on the Chat Message API as well as the global push toward larger context sizes opens an interesting opportunity to compare recent models in a perfectly equivalent setup on a reasoning-heavy dataset: MATH. Models We present an evaluation of the following models on the

Stay on top of key projects with a custom Dust assistant. From rolling out Dust and assistants across companies, we've seen tremendous value in creating AI assistants that provide project status updates. This best practice outlines how to build an assistant that delivers timely, relevant info on any

Take the load off your data analysts with a Dust assistant specialized in SQL. From wide-scale rollouts, we've seen immense value in AI assistants that automatically generate SQL queries. This article outlines building a Dust Custom assistant that outputs analysis-ready SQL. Simply include your most relevant table schemas

Based on our experience rolling out Dust assistants to augment our users’ customer service teams, we’ve compiled below some best practices to serve this use case better. Please reach out at team@dust.tt with any questions! We’ve repeatedly seen that a general-purpose assistant with access to the

Dust is an AI-native company based in Paris, France. We’re excited to share that we’ve raised €5M in seed funding led by Sequoia Capital with participation from XYZ, GG1, AIGrant, Connect Ventures, Motier VC, Remote First, Tiny Supercomputer, Seedcamp, and others, to build the operating system for smart

Behind the beautiful name of Speculative sampling lies a neat technique to have a large language model generate tokens up to three times faster 🔥 The technique has been developed by various research teams, including one from Google DeepMind who published it here. See for yourself: And also try it yourself

The opportunity opened by Large Language Models (LLMs) is vast. It’s allowed the birth of Dust, an AI-native company that wants to make work “work better” for smart teams. Deciding on a product constitution to guide what Dust would focus on in the vast space of possible directions was

Information overload is a millenia-old challenge that's only been amplified by digital technology. With Large Language Models (LLMs), we can leverage technology to overcome this challenge that every company faces. A new approach is needed: adaptive software that helps teams better harness information flows with software and free