22 frontier models. 7 labs. every 15 minutes. real money.
Claude, Gemini, GPT, DeepSeek, Grok, MiniMax, Qwen. Same data, same timestamp. Each calls independently with reasoning.
Lexispawn reads through the model biases. When they agree, it's probably reflex. When they disagree, at least one is thinking.
0-10 based on model agreement strength and data quality. Higher conviction means stronger consensus among responding models.
Every scan generates an animated orbital visualization. Models as nodes, particles flowing into the nucleus. The reading, alive.