GPT-6.1 Terra brings more advanced computer agent use at a lower cost than Claude
Affordability seems to be a big focus for OpenAI.
In a development that promises more advanced computer agent use at a lower cost, OpenAI has unveiled GPT-6.1 Terra. This new model, announced on the company's website, claims to outperform its predecessors while being more cost-effective. According to OpenAI, GPT-6.1 Terra has been tested using the Agent s Last Exam, a rigorous assessment of an agent's ability to execute complex, multi-step computer tasks accurately.
The results, displayed in a chart, show GPT-6.1 Terra's performance dominance across all effort levels, with the most powerful model (represented by the rightmost dot) significantly surpassing its counterparts. In comparison to Claude Opus 5.5 from Anthropic, GPT-6.1 Terra lags by just 3.2 percentage points, yet costs a fraction of the price at $1.01 compared to Claude's $8.82.
The new model is currently accessible to Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex, with a generous allowance of 1 million input tokens for $1 and 1 million output tokens for $5. While benchmarks offer a glimpse into an LLM's capabilities, the true test lies in real-world application.
Written by urgent.news from XDA Developers's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
