Aurora2-400M-A20M vs Aion 1.0 Instruct 2.3B
Compare documented capabilities and access. A dash means a comparable figure is unavailable.
| Specification | ||
|---|---|---|
| Task | Base language modeling | On-device language generation |
| Published capabilities | English base model pretrained on 100 billion tokens with one shared expert and 32 routed experts per layer. | Windows-oriented instruction model for local text generation, including document and email tasks. |
| Model size | 400M total / approximately 19M active parameters | 2.3B parameters |
| Access | Public Transformers checkpoint for completion and adaptation | Downloadable checkpoint |
| Operating limits | — | 32K-token context window |
| Sources | Model weights | Model weights |
| Inputs | Text | Text |
| Weights | Open weights · Apache-2.0 | Open weights · MIT |
| Released | Nov 18, 2026 | Nov 18, 2026 |