LFM2.5-Encoders for Fast Long-Context Inference on CPU Post published:2026-07-29 Read more articles Previous PostScientific computing in the age of agentic AI Next PostThe OlmoEarth Platform: Geospatial inference at planetary scale