Earth Virtual Expert (EVE) is an AI-driven Large Language Model fine-tuned for Earth observation and Earth science, developed to serve as a Digital Assistant for EO textual information.

Large Language Model for Earth Observation & Earth Science

Q: What are the objectives of EVE?

EVE aims to harness Earth Observation knowledge, launch an open-source LLM focused on EO, release an Earth Observation Virtual Expert, equip users with access to EO knowledge, and ensure compliant AI development.

An Earth Virtual Expert (EVE)

Find out more

Consortium members

EVE is funded by ESA Φ-lab and developed by Pi School in collaboration with Imperative Space.

Find out more

Objectives

The overall objective of this activity is to contribute to the development of a new generation of AI-powered 'Digital Assistant' interfaces for Earth Observation and Earth Science textual information. This will be achieved by:

Harness Earth Observation (EO) Knowledge

An icon showing the earth with a satellite orbiting

Launch an Open-source LLM focused on EO

A icon showing a computer with a brain which has the Earth inside

Release an Earth Observation Virtual Expert

Equip users with access to EO Knowledge

An icon showing a person with a text bubble with a satellite inside

Compliant AI Development

Find out more

Use Cases

Read about some of the potential use cases for EVE below

Beginner EO Concepts

EVE will help users understand basic EO terms & concepts

Advanced EO Concepts

Users can ask for explanations of more advanced EO concepts

EO Data Sources and Access Guidance

EVE will be able to provide information on accessing and downloading EO data

Summarisations

EVE will provide summarisations of EO scientific text

Quick access to EO insights

EVE will provide quick access to EO and Earth Science insights allowing users to ask critical questions

General conversational QA

EVE will be able to answer a wide variety of general questions from users

News

Discover the latest news and updates on the EVE project

EVE Presented at GEO-AI Working Group Webinar

February 20, 2026

The EVE team from Pi School, together with ESA’s Nicolas Longépé, presented the EVE project during a recent GEO Geospatial Artificial Intelligence Working Group webinar

View

Webinar: Introducing EVE, an Open-Source Earth Science LLM

January 30, 2026

Watch the Webinar replay and discover how EVE is making Earth Observation AI open and accessible.

View

EVE Featured in the AI Meets EO Education Webinar

January 29, 2026

Àlex R. Atrio presented EVE in a webinar, showing how AI can enhance EO training. The event, co‑hosted by EOTEC DevNet, EO College, and the Generative AI Community, gathered experts in EO education.

View

Large Language Model for Earth Observation & Earth Science

Consortium members

Objectives

Harness Earth Observation (EO) Knowledge

Launch an Open-source LLM focused on EO

Release an Earth Observation Virtual Expert

Equip users with access to EO Knowledge

Compliant AI Development

Use Cases

Beginner EO Concepts

Advanced EO Concepts

EO Data Sources and Access Guidance

Summarisations

Quick access to EO insights

General conversational QA

News

EVE Presented at GEO-AI Working Group Webinar

Webinar: Introducing EVE, an Open-Source Earth Science LLM

EVE Featured in the AI Meets EO Education Webinar

Subscribe to our Newsletter