List your company

Language Modeling on FineWeb-Edu (test)

16.46Perplexity (Test)Mamba-3-MIMO-180M
12.734437.882263.0388.1778Feb 11, 2026Mar 1, 2026Mar 19, 2026Apr 6, 2026Apr 24, 2026May 12, 2026May 31, 2026

Evaluation results

MethodLinks
Mamba-3-MIMO-180MParams=180M, Tokens=100B, Zero-shot=true2026.0516.46——————
Echo-180MParams=180M, Tokens=10B, Zero-shot=true2026.0516.48——————
GDN-180MParams=180M, Tokens=100B, Zero-shot=true2026.0516.52——————
Mamba-3-SISO-180MParams=180M, Tokens=100B, Zero-shot=true2026.0516.59——————
Mamba-2-180MParams=180M, Tokens=100B, Zero-shot=true2026.0516.76——————
Transformer-180MParams=180M, Tokens=100B, Zero-shot=true2026.0516.89——————
BaseBudget=24x, Configuration=(24 ⊗ 1), Params=24x, FLOPs=24x2026.0220.7——————
BaseBudget=12x, Configuration=(12 ⊗ 1), Params=12x, FLOPs=12x2026.0222.24——————
TMLTBudget=24x, Configuration=(3 ⊗ 8), Params=3x, FLOPs=24x2026.0222.87——————
LoopFormerBudget=24x, Configuration=(3 ⊗ 8), Params=3x, FLOPs=24x2026.0222.87——————
TMLT-EEBudget=24x, Configuration=(3 ⊗ 8), Params=3x, FLOPs=24x2026.0224.07——————
Base-LoopBudget=24x, Configuration=(3 ⊗ 8), Params=3x, FLOPs=24x2026.0224.53——————
DAG-MoE-lNumber of Parameters=699M2026.0524.69——————
LoopFormerBudget=12x, Configuration=(3 ⊗ 4), Params=3x, FLOPs=12x2026.0225.02——————
BaseBudget=6x, Configuration=(6 ⊗ 1), Params=6x, FLOPs=6x2026.0225.28——————
Base-Loop-EE-ConsBudget=24x, Configuration=(3 ⊗ 8), Params=3x, FLOPs=24x2026.0225.33——————
MoE-lNumber of Parameters=699M2026.0525.38——————
LLaMAContext Length=512, h=512, L=12, Ex=339k, scale=Large-scale2026.0225.8——————
MambaContext Length=512, h=512, L=12, Ex=339k, scale=Large-scale2026.0226.5——————
Naive-Loop-EEBudget=24x, Configuration=(3 ⊗ 8), Params=3x, FLOPs=24x2026.0226.55——————
Naive-Loop-EEBudget=12x, Configuration=(3 ⊗ 4), Params=3x, FLOPs=12x2026.0226.74——————
Echo-50MParams=50M, Tokens=3B, Zero-shot=true2026.0527.48——————
Base-Loop-EE-ConsBudget=12x, Configuration=(3 ⊗ 4), Params=3x, FLOPs=12x2026.0227.72——————
TMLT-EEBudget=12x, Configuration=(3 ⊗ 4), Params=3x, FLOPs=12x2026.0228.28——————
Naive-Loop-EEBudget=6x, Configuration=(3 ⊗ 2), Params=3x, FLOPs=6x2026.0229.36——————
ProtoTContext Length=512, h=512, L=12, Ex=339k, scale=Large-scale2026.0229.5——————
DeltaNetContext Length=512, h=512, L=12, Ex=339k, scale=Large-scale2026.0231.5——————
LoopFormerBudget=6x, Configuration=(3 ⊗ 2), Params=3x, FLOPs=6x2026.0233.45——————
Base-Loop-EE-ConsBudget=6x, Configuration=(3 ⊗ 2), Params=3x, FLOPs=6x2026.0235.95——————
TMLT-EEBudget=6x, Configuration=(3 ⊗ 2), Params=3x, FLOPs=6x2026.0237.83——————
LLaMAContext Length=2048, h=256, L=6, Ex=18k2026.0263.6——————
LLaMAContext Length=1024, h=256, L=6, Ex=18k2026.0265.3——————
DeltaNetContext Length=2048, h=256, L=6, Ex=18k2026.0268.9——————
MambaContext Length=2048, h=256, L=6, Ex=18k2026.0269.5——————
LLaMAContext Length=512, h=256, L=6, Ex=18k2026.0270.4——————
MambaContext Length=1024, h=256, L=6, Ex=18k2026.0270.5——————
DeltaNetContext Length=1024, h=256, L=6, Ex=18k2026.0270.6——————
ProtoTContext Length=2048, h=512, L=6, Ex=18k2026.0273——————
ProtoTContext Length=1024, h=512, L=6, Ex=18k2026.0273.7——————
ProtoTContext Length=1024, h=256, L=12, Ex=18k2026.0274.6——————
DeltaNetContext Length=512, h=256, L=6, Ex=18k2026.0276.3——————
ProtoTContext Length=2048, h=256, L=12, Ex=18k2026.0276.7——————
MambaContext Length=512, h=256, L=6, Ex=18k2026.0278——————
LLaMAContext Length=256, h=256, L=6, Ex=18k2026.0278.7——————
ProtoTContext Length=1024, h=256, L=6, Ex=18k, R=642026.0279.1——————
ProtoTContext Length=512, h=256, L=12, Ex=18k2026.0279.8——————
ProtoTContext Length=1024, h=256, L=6, Ex=18k2026.0280.5——————
ProtoTContext Length=2048, h=256, L=6, Ex=18k, R=642026.0280.5——————
ProtoTContext Length=2048, h=256, L=6, Ex=18k2026.0281.9——————
ProtoTContext Length=512, h=512, L=6, Ex=18k2026.0282——————
ProtoTContext Length=512, h=256, L=6, Ex=18k, R=642026.0283.2——————
ProtoTContext Length=512, h=256, L=6, Ex=18k2026.0284.8——————
MambaContext Length=256, h=256, L=6, Ex=18k2026.0286——————
DeltaNetContext Length=256, h=256, L=6, Ex=18k2026.0290.4——————
ProtoTContext Length=256, h=256, L=6, Ex=18k2026.0290.5——————
ProtoTContext Length=256, h=256, L=6, Ex=18k, R=642026.0294——————
ProtoTContext Length=256, h=512, L=6, Ex=18k2026.0297.2——————
ProtoTContext Length=256, h=256, L=12, Ex=18k2026.02109.6——————
MelaParams=400M2026.05—12.5312.7512.0112.6414.4314.5
MelaParams=800M2026.05—10.510.7510.0410.8212.7413.41
MelaParams=1.2B2026.05—9.089.499.29.9111.6411.85
Transformer++Params=400M2026.05—13.5914.0212.5628.26130.21303.56
Transformer++Params=800M2026.05—12.1712.4211.3516.69129.1497.85
Transformer++Params=1.2B2026.05—10.2710.469.5312.71104.14597.37