“Within the next 18 months, you will be able to host GLM 5.2 equivalent intelligence on an RTX 5090 GPU.” -Ahmad Osman, AI World’s Fair

Ahmad Osman stated at AI World's Fair that within 18 months, GLM 5.2-equivalent AI intelligence will be deployable locally on a single RTX 5090 GPU, indicating rapid progress toward running advanced language model capabilities on consumer-grade hardware. This prediction suggests significant improvements in model efficiency and optimization techniques that would enable previously cloud-dependent AI workloads to run on personal computing devices.

Source: Swyx (X) | 2026-07-02