October 27, 2026 · InfoWorld
OpenAI bets on AI economics with lower-cost GPT-6.1 Terra
My take: OpenAI just launched GPT-6.1 Terra, a model built for complex tasks (visual reasoning, computer use, professional work) at a fraction of the cost of its pricier models. By OpenAI's own testing, Terra solves agentic tasks with a 60% success rate at just over a dollar per task, compared to the more than seven dollars its flagship model, Astra, costs for a barely better result.
This is exactly the direction that matters for any business already using AI for real work: it is not about having the most powerful model on the market, but the one that gives you the best result for every dollar you spend. Since OpenAI ran these tests itself, comparing its own model against the competition (including Claude), it is worth reading those numbers with your own judgment until someone independent confirms them.
For anyone building with AI every day, like me, the real question is never "which model is the best?" but "which one gives me what I need at the lowest possible cost?".
If your business already uses AI in its workflow, have you compared the cost per task of the tools you use, or do you just go by which one sounds most advanced?
Want to use these tools? See the unbiased reviews or back to the news.