on-device

taxonomy axis: design_intent

Decision rule: Framed for edge/deployment efficiency constraints.

Anchor: Phi-3-mini

Models in this category (4)

ModelOrgDateBlockAttentionContextTotalActive
Phi-3Microsoft2024-04-22denseMHA4,0963.8B3.8B
Gemma 3Google DeepMind (Gemma Team)2025-03-25densesliding-window131,07227B27B
Tiny AyaCohere Labs / Cohere2026-03-12densehybrid8,1923.35B3.35B
Gemma 4Google DeepMind (Gemma Team)2026-07-24densehybridnot disclosed31.25B31.25B

What varies within this category

attention variant varies — MHA, hybrid, sliding-window. trained context varies — 131072, 4096, 8192, not disclosed. openness varies — open-weights, undisclosed. scale class varies — large, medium.