EdgeRunner AI and the US Army’s Artificial Intelligence Integration Center have developed EdgeRunner-Camo, an open-weight large language model designed specifically for military applications. The model reportedly reduces error rates by up to 37% across selected Army tasks. Fine-tuned for local deployment, EdgeRunner-Camo can operate in air-gapped environments without depending on cloud-based AI services. Its development was based on analysis of 1.4 million CamoGPT conversations containing 17.7 million messages. The teams filtered the data to remove non-military queries, personally identifiable information, and questions requiring external tools or retrieval. The approach is intended to provide the Army with a more reliable AI capability that can operate securely in environments where cloud connectivity is unavailable or restricted.
