EXPLORINEER
en

Archive — US — techcrunch

OpenAI introduces ‘Ultrafast,’ a new mode that makes GPT-5.6 Sol work at 14x thespeed

OpenAI has launched 'Ultrafast,' a new mode for its GPT-5.6 Sol model that operates at 14 times the speed of standard processing, generating up to 750 output tokens per second. This mode enhances performance in workflows like customer service and financial analysis, leveraging a partnership with Cerebras for chip support, and is currently available in preview for select customers.

Key points

  • OpenAI released a new processing mode named 'Ultrafast' for GPT-5.6 Sol.
  • Ultrafast operates at 14x the speed of regular processing.
  • It can generate up to 750 output tokens per second.
  • The mode is aimed at enhancing corporate workflows such as incident response and customer support.
  • The introduction of Ultrafast represents a shift toward more efficient AI outputs.
  • Competitors like Anthropic offer accelerated models, but they don't match Ultrafast's speed.
  • Cerebras chips power this feature, indicating a significant hardware collaboration.
  • Currently, access is limited to a small group of customers with plans for broader availability.

Why it matters

  • The trend of increasing AI processing speed aligns with the ongoing demand for real-time data processing in various industries.
  • Faster AI models enhance productivity, which is crucial for businesses aiming for efficiency.
  • The AI industry's competitive landscape is marked by rapid advancements, with companies like OpenAI and Anthropic innovating to capture market share.
  • The collaboration with chip manufacturers indicates a growing trend of hardware-software synergy in AI development.

OpenAI's Ultrafast mode for GPT-5.6 Sol significantly raises AI performance benchmarks, aligning with industry trends toward faster and more efficient technology, setting a crucial standard for future AI capabilities.