ZMIME

Aurora2-400M-A20M vs Olala Mini 7A1B

Compare documented capabilities and access. A dash means a comparable figure is unavailable.

Model specifications
SpecificationAurora2-400M-A20MAuroraAI ResearchOlala Mini 7A1BOVHcloud
TaskBase language modelingMultilingual language generation
Published capabilitiesEnglish base model pretrained on 100 billion tokens with one shared expert and 32 routed experts per layer.Compact mixture-of-experts model combining hybrid attention and Mamba layers.
Model size400M total / approximately 19M active parameters7B total / 1B active
AccessPublic Transformers checkpoint for completion and adaptationDownloadable checkpoint for local inference or fine-tuning
SourcesModel weightsModel weights · Additional documentation
InputsTextText
Context window—32,000 tokens
WeightsOpen weights · Apache-2.0Open weights · Apache-2.0
ReleasedNov 18, 2026Nov 19, 2026