Live demo shows swapping harness components and system prompts shifts model performance

The author released an interactive HF Space demo letting users tweak system prompts and tools to see real-time performance changes, and continued experiments showing that even simple system prompt modifications significantly affect model performance.

2026-10-07 ~ 2026-10-07 · 2 related posts

• Swapping just the system prompt measurably changes harness benchmark results — yb2698 · 2026-10-07
• Interactive demo shows how changing harness components affects model task solving — yb2698 · 2026-10-07