skip to content
vortx.ai
Add @emem
Vortx/Research

The work behind the protocol.

  • readthe papers on Zenodo
  • runthe models on Hugging Face and the protocol from GitHub, Apache-2.0
  • checkevery claim against its live endpoint, then decide for yourself
read 2026-06-15 emem: A Content-Addressed, Verifiable Earth-Memory Protocol for AI Agents over Foundation-Model Embeddings
read 2026-06-15 emem: A research on Content-Addressed, Verifiable Earth-Memory Protocol for AI Agents over Foundation-Model Embeddings
run model TerraGround-Gemma-4-12B-LoRA
run model TerraGround-Gemma-4-E2B-LoRA
fork source github.com/Vortx-AI/emem (the protocol), Apache-2.0

Reproduce it, live.

run the papers’ claims as public endpoints, from your terminal: no key, no account.

What the responder can do today

terminal
curl -s https://emem.dev/v1/capabilities   # models loaded, GPU state, live extensions

The transparency log's signed tree head

terminal
curl -s https://emem.dev/v1/log/sth   # every fact lands in a witnessed, Merkle-proofed log

The benchmark the papers describe

terminal
curl -s https://emem.dev/v1/benchmark   # run the numbers yourself, not our summary of them

Why we publish

  • citea fact only if you can check it; so the method is open too: the design written up, the models downloadable, the code readable
  • findanything wrong here and it can be corrected, in the open
  • readthe papers by Jaya Kumari and Avijeet Kumar Singh, on Zenodo under a Creative Commons licence: 10.5281/zenodo.20706317 and 10.5281/zenodo.20706893
  • runthe models, LoRA fine-tunes of Gemma that ground a language model in Earth observation, on the protocol itself, emem