Thinking Machines Unveils ~1T Omni Model with 1M Context Window and Multi-modal Training

Core summary

One sentence to understand this update

Thinking Machines has released a ~1T Omni model featuring a 1M context window, trained on 48T tokens across image, text, and audio data.

Impact & opportunity

What this could mean

This powerful multi-modal model with an immense context window opens new avenues for developers to build highly capable and context-aware AI applications.