Cleveris Research

Advancing the frontier of artificial intelligence through open research and collaborative innovation.

Cleveris Releases

# Project Name Type Details Release Date Size
0 LV Agent Agent An open-source terminal agent built on the Karpathy philosophy — event-sourced harness, real semantic memory, deep research, and self-evolving skills. It decides fast vs. deep, calls 18+ tools, generates PDFs, and understands your conversation context. 08/17/25 N/A
1 Cleveris Agent Agent Under development. 04/25/25 N/A
2 Cleveris-7B Model Under development. 04/20/25 14 GB
3 Cleveris Math 1 Model Under development. 04/15/25 60 GB
4 Cleveris-4.5-Base-38B Model Under development. 04/10/25 76 GB
5 Cleveris 4 Technical Report Paper Under development. 04/05/25 N/A
6 Measuring Thinking Efficiency in Reasoning Models Research Under development. 03/20/25 N/A
7 Cleveris 3 Dataset Dataset Under development. 03/15/25 N/A
8 Sequential Monte Carlo for LLMs Research Under development. 03/10/25 N/A
9 Cleveris Network Training Under development. 03/05/25 N/A
10 Atropos Framework Framework Under development. 03/01/25 N/A

About Cleveris Research

Cleveris Research is dedicated to advancing artificial intelligence through open research, collaborative development, and transparent publication of our findings. We believe that AI should be developed in the open, with contributions from researchers worldwide.

Our work spans foundation models, reasoning systems, autonomous agents, and the infrastructure needed to train them at scale. We publish our models, datasets, and research papers to accelerate progress in the field.

Featured — LV Agent: Our flagship terminal agent embodies minimal, verifiable engineering. It runs a capability kernel with policy-as-data security, real vector semantic memory, adaptive fast-vs-deep routing, iterative deep research with evidence cross-validation, and skills that evolve into first-class tools. Built to be understood, not just to run.