# Edouard Grave

Edouard Grave is a French computer scientist and co-founder of Mistral AI, specializing in machine learning and large language models. He formerly led research teams at Meta AI and holds a PhD from Sorbonne University.

Edouard Grave is a French researcher in [machine-learning](https://www.wikiprompt.org/wiki/machine-learning) and a co-founder of [large language model](https://www.wikiprompt.org/wiki/large-language-model) company Mistral AI. His work centers on [neural networks](https://www.wikiprompt.org/wiki/neural-network), [deep-learning](https://www.wikiprompt.org/wiki/deep-learning), and scalable [transformer](https://www.wikiprompt.org/wiki/transformer) architectures, with contributions to both academic research and industry deployment.

Grave earned his PhD from Sorbonne University in Paris, where his doctoral research focused on natural language processing and representation learning. Before founding Mistral AI in 2023, he spent several years at Meta AI (formerly Facebook AI Research) in Paris, leading teams working on self-supervised learning and efficient [neural](https://www.wikiprompt.org/wiki/neural-network) models. He also held research positions at institutions such as [nokia-bell-labs](https://www.wikiprompt.org/wiki/nokia-bell-labs) and the École Normale Supérieure.

## Mistral AI and founding

In 2023, Grave co-founded Mistral AI with fellow researchers Arthur Mensch and Timothée Lacroix. The company launched with a focus on open-weight [large language models](https://www.wikiprompt.org/wiki/large-language-model), positioning itself as a European alternative to US-based labs like [openai](https://www.wikiprompt.org/wiki/openai) and [anthropic](https://www.wikiprompt.org/wiki/anthropic). Mistral's first model, Mistral 7B, was released in September 2023 and gained attention for its performance relative to parameter count.

Grave's role at Mistral involves guiding research on model architecture and training methods. The company released subsequent models including Mixtral 8x7B, a sparse mixture of experts model, and the larger Mistral Large, targeting enterprise clients through cloud providers such as [azure](https://www.wikiprompt.org/wiki/azure), [google-cloud](https://www.wikiprompt.org/wiki/google-cloud), and [amazon-web-services](https://www.wikiprompt.org/wiki/amazon-web-services). Mistral secured significant venture funding, reaching a valuation of several billion euros by 2024.

## Research contributions

Grave's academic work includes studies on word embeddings, [transformer](https://www.wikiprompt.org/wiki/transformer) efficiency, and [self-supervised learning](https://www.wikiprompt.org/wiki/self-supervised-learning). He co-authored papers on adaptive word representations and methods for reducing the computational cost of [neural](https://www.wikiprompt.org/wiki/neural-network) language models. His research influenced later work on efficient [LLMs](https://www.wikiprompt.org/wiki/large-language-model), particularly in contexts with limited hardware resources.

At Meta AI, Grave worked on projects integrating [machine-learning](https://www.wikiprompt.org/wiki/machine-learning) with multimodal data, including text and image understanding. He contributed to tools and benchmarks that shaped [generative-ai](https://www.wikiprompt.org/wiki/generative-ai) development during the late 2010s and early 2020s.

## Public engagement

Grave has spoken at conferences and workshops on [artificial-intelligence](https://www.wikiprompt.org/wiki/artificial-intelligence) policy and open-source models. He has advocated for transparency in AI development, arguing that open-weight models enable broader research access and innovation. His public statements have addressed concerns about [AI safety](https://www.wikiprompt.org/wiki/ai-safety) and the concentration of power among a few large technology firms.

## Recognition and impact

As a co-founder of Mistral AI, Grave has been featured in industry press and technology rankings. The company's rapid growth made it one of the most prominent AI startups in Europe, with partnerships and deployments across multiple sectors. Grave's work bridges academic theory and practical systems, contributing to advances in [deep-learning](https://www.wikiprompt.org/wiki/deep-learning) and [natural language processing](https://www.wikiprompt.org/wiki/natural-language-processing).

His career reflects a trajectory from foundational research at academic institutions to leading a commercial [LLM](https://www.wikiprompt.org/wiki/large-language-model) venture, positioning him as a notable figure in the current wave of [generative-ai](https://www.wikiprompt.org/wiki/generative-ai).

---
Source: https://www.wikiprompt.org/wiki/edouard-grave
License: CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/)
Last updated: 2026-09-07T21:29:14.924888+00:00