https://smartchunks.com/zgcm-1-open-7b-foundation-model-efficient-math-agentic-search/Smart Chunksen2026-09-15T09:04:25-04:00ZGCM-1: The Fully Open 7B Model Betting Efficiency Beats Scale https://smartchunks.com/gauge-llm-judge-ai-agent-evaluation-flaws/Smart Chunksen2026-09-14T09:06:12-04:00GAUGE Study Finds LLM Judges Can’t Tell Which AI Agents Actually Finish the Job