Controversy over unreleased experimental Llama 4 Maverick behind Arena score
Controversy arose because an unreleased experimental version was used for Llama 4 Maverick's Arena score. Meta denied allegations of training on test sets and attributed the issue to deployment timing
Why it matters
The mismatch between versions submitted to leaderboards and the publicly released ones became a public issue
Source — Primary sources first; secondary reports are labelled “Press”
Presstechcrunch.comhttps://techcrunch.com/2025/04/07/meta-exec-denies-the-company-artificially-boosted-llama-4s-benchmark-scores/Added in Week 35, 2026 · See that week