Chat with the combined power of AI models from OpenAI, Google, XAI, Anthropic and more, to achieve results 5.5% better than the best model in the world.
SuperVessel combines the power of every LLM, outperforming the current best model in the world by 5.5% on the Humanity's Last Exam benchmark.
The unified intelligence of all models working in perfect harmony
SuperVessel performed 5.5% stronger relative to the previous best model, Gemini 3 Pro, both on our internal comparison and on the publically reported figures by Google.
SuperVessel
Gemini 3 Pro
GPT 5.2
Claude 4.5 Opus
Grok 4.1