Key facts
- Google released Gemini 3.7 Flash, a new AI model for coding and autonomous workflows.
- Gemini 3.7 Flash offers improved performance in coding and complex document processing compared to its predecessor.
- OpenAI previewed GPT-5.6 Sol Ultrafast, a faster tier of its existing model powered by Cerebras chips.
- Gemini 3.7 Flash is generally available, while OpenAI's Ultrafast tier is currently invite-only.
- Google is emphasizing cost-effectiveness for Gemini 3.7 Flash with introductory pricing.
Google has launched Gemini 3.7 Flash, its latest AI model, just three weeks after releasing Gemini 3.6 Flash. The new model is designed for coding and autonomous business workflows and is now generally available. This rapid release aims to bolster Google's position in the competitive AI landscape, particularly in the race for faster and more efficient AI agents.
OpenAI simultaneously previewed GPT-5.6 Sol Ultrafast, a service tier that runs its most capable model significantly faster, powered by Cerebras wafer-scale chips. While Google's Gemini 3.7 Flash is available to all developers, OpenAI's Ultrafast tier is currently in a limited preview.
Google claims Gemini 3.7 Flash offers substantial performance gains in coding and complex document processing, citing benchmark scores that surpass its predecessor. The company is also emphasizing the model's cost-effectiveness with introductory pricing, positioning it as a competitive option for autonomous systems. OpenAI has not released detailed head-to-head benchmark scores for its Ultrafast tier but has shared customer testimonials highlighting its speed and utility for real-time applications.
The push for speed in AI development signals a shift from focusing solely on raw intelligence to enabling real-time agent capabilities. Google's release comes amidst speculation about its development pace, particularly concerning its flagship Gemini 3.5 Pro model, while OpenAI is leveraging external chip technology to accelerate its offerings.