How Bella is trained and what data it uses

Bella is powered by a combination of multiple state-of-the-art AI models via the OpenRouter API, including Claude Opus 4.8 (Anthropic), ChatGPT 5.2 (OpenAI), Gemini 2.5 Pro (Google), and others. You can choose which model powers your conversations from the model selector in chat.

What makes Bella unique is not the underlying AI model but the additional layers built on top of it:

1. Human Design knowledge layer

The HumanDesign.ai platform supplies each AI model with a deep, structured knowledge base of Human Design material: gate and line descriptions, channel dynamics, type strategies, authority types, profiles, variables, incarnation crosses, and teachings from Ra Uru Hu. General AI models do not have this level of Human Design depth on their own.

2. Your personal chart data

Bella receives your actual Bodygraph data with every message: all 1,000+ data points including activated gates and channels, defined and undefined centres, profile, type, authority, and more. This is what allows Bella to give personalised answers rather than generic Human Design explanations.

3. Chart context and memory

Bella can use Chart Context notes you add for any chart, plus Bella Memory (saved context from previous conversations), to further personalise responses over time.

Accuracy

With the above layers, Bella achieves around 98-99% accuracy when reading Bodygraph data. Accuracy is continuously improved through updates to the knowledge base and system prompts across all supported AI models.