Paris-based Dottxt raises €10.9M to help AI speak the language of every application; here’s how

Paris-based Dottxt, an ecosystem and platform for interacting with large language models (LLMs), has raised $11.9M (approximately €10.97M) in total funding through pre-seed and seed rounds completed within just over seven months.

The funding is split into a $3.2M pre-seed round led by Elaia in December 2023 and an $8.7M seed round led by EQT Ventures in August 2024. 

Investors include Seedcamp, Common Magic, Kima, FSJ, Roxanne Varza (Station F), Erik Bernhardsson (Modal Labs, CEO), Julien Chaumond (Hugging Face, CTO), Bob van Luijt (Weaviate, CEO) and Jean-Louis Queguiner (Gladia, CEO).

Julien Hobeika, Partner at EQT Ventures, says, “From our first discussion with Dottxt, we saw their potential to turn the stochastic nature of LLMs into something more programmatic and deterministic, enabling deep integration into software systems.”

“This aligns with our belief that the future of AI agents will be shaped by those who can build the essential frameworks that make LLMs truly usable at scale in critical enterprise systems. We’re excited to support Dottxt as they lead the creation of this critical infrastructure.”

Making AI speak computers

Dottxt enhances large language models (LLMs) by enabling users to request structured information, making LLMs more than just conversational tools. This upgrade allows seamless integration with digital ecosystems, transforming LLMs into reliable computational resources.

Data scientists can issue natural language database queries with accuracy, while HR professionals can filter CVs for specific qualifications. Additionally, users can quickly extract desired attributes from large image banks, streamlining workflows and saving time across various industries.

Co-founders Rémi Louf, Dan Gerlanc and Brandon Willard, who have collaborated for over four years, developed Dottxt while working at the New York-based AI firm Normal Computing. They encountered challenges with GPT-4, particularly in systematising data extraction to avoid manual work. Leveraging their expertise in statistical modeling and compiler technology, they devised a solution no one else had found.

In just a year, Dottxt’s open-source library, Outlines, has garnered over 3 million downloads, including 600,000 in the past month. It is being used in production by companies of all sizes, with major AI players like OpenAI and Cohere among the early adopters.

Capital utilisation

The funds will be used to expand Dottxt’s team, increase the number of software engineers and hire a Chief of Staff. The team, currently at nine members, has added six new employees to focus on extending the technology’s reach and distributing it to enterprises and developers.

Rémi Louf, CEO at Dottxt, says, “Everyone will be using structured generation in a few years, there is no doubt about that. Model providers, including OpenAI, are lagging in terms of speed and capabilities, and we’re here to fill that gap.”

“With these funds, we will keep pushing the limits of this technology and make it more widely available for everyone. We are shaping a future where generative AI delivers on the kind of automation we were promised.”

What do the investors have to say?

Clement Vanden Driessche, Investment Director at Elaia, says, “In a new software era led by the emergence of AI agents, developers experiment with a strong need to constrain the LLM models’ behaviour to make the final applications faster, robust, and reliable.”

“At Elaia, we are proud to support Dottxt team, from their inception and even before, who developed a proprietary technology inspired by compilers. We are excited to see them grow in the open-source ecosystem with an incredible developers adoption to date.”

Sia Houchangnia, Partner at Seedcamp, adds, “LLMs already have the potential to unlock tremendous value, but their lack of reliability has been the key barrier to wider adoption. Dottxt solves this problem.”

“By applying their expertise in Bayesian statistics, Remi, Brandon, Dan, and their team have created the most accurate and high-performing platform for structured generation. We are proud to be day-one backers of the company and are convinced that Dottxt has the potential to become the default framework for LLM programming.”