27 August 2026

Perplexity and Nvidia release local AI agent software

First reported

AI Business ran this on , a day before the next source picked it up.

  • Perplexity, an AI search company, and Nvidia, a chip maker, jointly built software that runs AI agents on Nvidia's own hardware without paying per API call.
  • The system runs locally on DGX Spark and RTX devices, meaning computations happen on the user's machine rather than remote servers.
  • The agent software uses a 27-billion-parameter model, a medium-sized AI system, and reportedly outperforms similar open-source alternatives in tests.

How it was covered

Deep Learning WeeklyEditorial team

Perplexity and Nvidia co-designed a fully local AI agent stack on DGX Spark and RTX hardware with zero token costs. The newsletter emphasises the minimal harness outperforms open-source alternatives on the same 27B model.

Reported by AI Business