← Back to leaderboard

gemma4:e4b

3.4 GB measured VRAM · sub-4GB tier · rag-gauntlet-v1
Exact passages, questions, 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
227 tokens · 10.986s (representative seed)
Tier 2 — restraint (refusal) 5/5
306 tokens · 2.602s (representative seed)
Tier 3 — distractor resistance 5/5
266 tokens · 2.241s (representative seed)
Tier 4 — multi-passage synthesis 5/5
307 tokens · 2.5s (representative seed)
Tier 5 — precise extraction 5/5
43 tokens · 0.647s (representative seed)
Tier 6 — long-context needle 5/5
219 tokens · 1.968s (representative seed)