Comparative Ranking on Unified Evaluation v1 (aggregate)
1.81Average RankFEM-SM
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FEM-SMParameters=340M, Tokens=15B, Architecture Components=SM-Attn+C,L,T,G2026.02 | 1.81 | 8 | |
| FEM-SMParameters=1.3B, Tokens=100B, Architecture Components=SMAttn+C,L,T,G2026.02 | 2.06 | 9 | |
| FEM-SM(-G)Parameters=340M, Tokens=15B, Architecture Components=SMAttn+C,L,T2026.02 | 3.31 | 2 | |
| GSAParameters=1.3B, Tokens=100B2026.02 | 3.38 | 2 | |
| FEM-GLAParameters=1.3B, Tokens=100B, Architecture Components=GLA+C,L,T,G2026.02 | 3.88 | 1 | |
| GatedDeltaNetParameters=340M, Tokens=15B2026.02 | 4.25 | 1 | |
| DiffTransParameters=340M, Tokens=15B2026.02 | 4.38 | 1 | |
| DeltaNetParameters=1.3B, Tokens=100B2026.02 | 4.44 | 1 | |
| Transformer (SMAttn)Parameters=1.3B, Tokens=100B2026.02 | 4.56 | 1 | |
| FEM-SM(-G,T)Parameters=340M, Tokens=15B, Architecture Components=SMAttn+C,L2026.02 | 4.81 | 1 | |
| DeltaNetParameters=340M, Tokens=15B2026.02 | 5.44 | 3 | |
| GLAParameters=1.3B, Tokens=100B2026.02 | 5.63 | 0 | |
| HGRNParameters=1.3B, Tokens=100B2026.02 | 5.75 | 0 | |
| FEM-SM(-G,T,L,C)Parameters=340M, Tokens=15B, Backbone=SMAttn2026.02 | 6.5 | 1 | |
| FEM-GLAParameters=340M, Tokens=15B, Architecture Components=GLA+C,L,T,G2026.02 | 6.56 | 2 | |
| HGRN2Parameters=1.3B, Tokens=100B2026.02 | 6.63 | 2 | |
| FEM-SM(-G,T,L)Parameters=340M, Tokens=15B, Architecture Components=SMAttn+C2026.02 | 6.63 | 0 | |
| RetNetParameters=1.3B, Tokens=100B2026.02 | 7.63 | 0 | |
| GLAParameters=340M, Tokens=15B2026.02 | 9.38 | 0 |