NVIDIA's SoL-Pi Cuts Coding Agent Token Traffic ~49% and API Cost by a Third

nvidia · hf · 2026-09-18

NVIDIA introduces SoL-Pi, an RSI-inspired approach that recursively scales auto-research loops at the harness layer, letting agents discover reusable harness improvements across increasingly diverse environments that transfer beyond their development setting. Four mechanisms survive selection, spanning action execution, context compaction, observation handling, and delegated reading.

On the 51-task EdgeBench evaluation, SoL-Pi matches Pi's performance across GPT-5.6 Sol and Opus 5 while:

The work demonstrates that automated harness discovery via self-improving research loops can reach production-level outcomes.

Original post →

More from coding & agent

coding & agent channel →