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