← Back to leaderboard

gemma4:12b

8.4 GB measured VRAM · 8-12GB tier · reasoning-gauntlet-v1
Exact problems, puzzles, and expected answers for this category are held back from publication to avoid training-data contamination — see About for why. This page shows per-tier pass/fail counts and timing only.
Tier 1 — floor 5/5
333 tokens · 7.452s (representative seed)
Tier 2 — multi-step 5/5
1017 tokens · 12.256s (representative seed)
Tier 3 — logical deduction 5/5
549 tokens · 6.998s (representative seed)
Tier 4 — adversarial (CRT trap) 5/5
301 tokens · 3.995s (representative seed)
Tier 5 — combinatorics 5/5
623 tokens · 7.824s (representative seed)
Tier 6 — distractor filtering 5/5
503 tokens · 6.372s (representative seed)