OpenAI has introduced an Ultrafast mode for its GPT-5.6 Sol model, significantly enhancing inference speed. Powered by hardware from Cerebras, this new mode can generate outputs up to 14 times faster, reaching speeds of 750 output tokens per second, according to The Decoder.
This development is part of a broader $10 billion partnership between OpenAI and Cerebras, aimed at accelerating AI processing capabilities. The integration of Cerebras’ advanced hardware solutions is a key factor in achieving this substantial performance improvement.
For Japanese markets focused on FX, crypto, and equities, faster AI inference could translate into more timely data analysis and decision-making tools, potentially impacting algorithmic trading and real-time market insights.