OpenAI announced updates to its GPT-5.6 models, including price cuts and efficiency improvements. The company emphasized its focus on making advanced AI more affordable and widely accessible. These changes are part of a broader strategy to enhance the economic value of AI by reducing costs and improving performance. The updates reflect OpenAI's commitment to expanding the practical applications of AI across various industries and user groups. Source: openai
The pricing changes for GPT-5.6 Luna and Terra are a key part of OpenAI's efforts to make AI more accessible. Luna's price was reduced by 80%, bringing it to $0.20 per million input tokens and $1.20 per million output tokens. Terra's price dropped by 20%, costing $2 and $12 respectively. These adjustments aim to lower the barriers to entry for businesses and developers, enabling more widespread use of AI technologies. Additionally, GPT-5.6 Sol's Fast mode offers up to 2.5 times the speed of standard processing at twice the price, without compromising on the model's intelligence. Source: openai
OpenAI highlighted the importance of efficiency in its AI infrastructure, noting that improvements in speculative decoding increased token-generation efficiency by over 15%. The company also mentioned that better routing and context management help keep hardware productive and prevent agents from repeating work. In a recent benchmark analysis, enhancements to retained reasoning and context management boosted GPT-5.6 Sol’s score on the public ARC-AGI-3 task set from 13.3% to 38.3% while using six times fewer output tokens. These gains demonstrate the effectiveness of optimizing the surrounding system rather than just the model itself. Source: openai
