Aurora One
Aurora One is a compact experimental chat model designed for general conversation, lightweight assistance, and research.
Smartness
Aurora One is designed to:
- hold basic conversations
- answer straightforward questions
- explain simple concepts
- help with brainstorming and writing
- follow short instructions
- generate short stories and responses
- assist with basic coding and technical questions
Aurora One is still experimental. It may make factual mistakes, misunderstand complex instructions, repeat itself, or produce incorrect reasoning.
Intended Uses
Aurora One is intended for:
- general-purpose chat
- lightweight personal assistants
- educational experiments
- writing and brainstorming
- simple question answering
- basic coding assistance
- chatbot prototypes
- small-model research
- local and low-resource inference
It should not be relied on for medical, legal, financial, or other high-stakes decisions.
Model Parameters
| Property | Value |
|---|---|
| Parameters | 119,953,152 |
| Architecture | Llama-style causal transformer |
| Transformer layers | 14 |
| Hidden size | 768 |
| Attention heads | 12 |
| Key-value heads | 12 |
| MLP intermediate size | 2,304 |
Vocabulary
Aurora One uses a custom tokenizer with a vocabulary size of:
16,384 tokens
Context Length
Aurora One supports a maximum context length of:
2,048 tokens