OpenAI Announces New Model GPT-6.1 Terra with Visual Reasoning Capabilities

OpenAI has introduced its new artificial intelligence model, GPT-6.1 Terra, which delivers significant improvements in visual reasoning and computer-based task automation. Serving as the successor to GPT-5.6 Terra, this model stands out for its cost-efficiency alongside increases in speed and capacity.

GPT-6.1 Terra demonstrates strong performance results in benchmarks such as Agents' Last Exam, MMMU-Pro, and OSWorld 2.1. Achieving a 50.3 percent success rate on Terminal-Bench 4.0 in software engineering tasks, the model shows a notable decrease in error rates compared to the previous version, offering higher accuracy.

Providing a 50 percent discount in API pricing compared to the GPT-5.6 Terra and GPT-6.1 Sol models, the new model is priced at $1 per million input tokens, $0.05 per cached input token, and $5 per output token. The model has been released for access via ChatGPT Plus, Pro, Enterprise, and Edu subscribers as well as through the OpenAI API.