#agentic benchmark

Agentic Coding Benchmark Goes Open Source: A Deep Dive into the Hugging Face Dataset and Leaderboard

LeanZero open-sources an agentic coding benchmark to Hugging Face: 94 model records, per-round build scores, inference costs. We break down the Forge 2.0 leaderboard's Jev Router vs. GPT-6.1 Sol gap, the data's limitations, and why it matters.