OpenAI Launches GPT-6.1 Terra: Faster and Cheaper AI Model for Everyday Tasks

OpenAI has introduced GPT-6.1 Terra, a model designed for everyday professional tasks, offering faster performance, lower API costs, improved accuracy and enhanced computer-use capabilities.

OpenAI Introduces GPT-6.1 Terra With Faster Performance and Lower Costs

OpenAI has introduced GPT-6.1 Terra, a new artificial intelligence model designed to handle routine professional tasks more efficiently. According to the company, the model offers faster processing, lower operating costs and improvements in accuracy, coding and desktop application use compared with its predecessor, GPT-5.6 Terra.

The release focuses on practical workplace applications, including document creation, technical analysis and software interaction. OpenAI describes the model as a significant upgrade, combining visual understanding with computer-use capabilities and more natural writing.

GPT-6.1 Terra Designed for Visual and Computer-Based Tasks

GPT-6.1 Terra is built to support tasks that require interpreting visual information and interacting with computer applications. Its capabilities are intended to help professionals complete a variety of everyday assignments with less manual effort.

Potential applications include drafting business proposals, analysing technical drawings, filling out digital forms and editing video content.

According to OpenAI, internal professional evaluations also tested the model on more complex assignments. These included generating 3D models from architectural drawings and reviewing animated films to create reports containing time-coded observations.

The combination of visual analysis, software interaction and writing capabilities is intended to make the model useful across a wider range of professional workflows.

GPT-6.1 Terra Benchmark Performance and Speed

OpenAI has reported improvements in benchmark performance, task completion time and cost efficiency compared with earlier models.

15-Percentage-Point Improvement on Agents' Last Exam

On the Agents' Last Exam benchmark, OpenAI says GPT-6.1 Terra scored 15 percentage points higher than GPT-5.6 Terra when both were evaluated at medium reasoning effort.

The company also reported that the new model cost approximately one-third as much for the benchmark tasks.

Additionally, OpenAI said GPT-6.1 Terra completed tasks more than three times faster than GPT-6 Astra and GPT-6.1 Sol in the comparisons it described.

These results suggest potential benefits for users who regularly run AI-assisted workflows. However, benchmark results may differ from performance on individual real-world tasks.

Improved Accuracy and Fewer Factual Errors

Reducing incorrect responses is another area where OpenAI reports progress.

On the company's internal benchmark, GPT-5.6 Terra recorded an error rate of 53.6% at low reasoning effort. For GPT-6.1 Terra, the reported rate fell to 16.7% at the same effort level.

OpenAI said the error rate dropped further to 4.3% when the new model operated at maximum reasoning effort.

The results indicate a substantial improvement in the company's reported evaluation. Actual accuracy can still vary depending on the task, the information provided and the complexity of the request.

GPT-6.1 Terra Coding Performance

OpenAI also highlighted improvements in software development tasks.

On the Terminal-Bench 4.0 coding benchmark, GPT-6.1 Terra solved 50.3% of tasks at maximum reasoning effort. The corresponding published score for GPT-5.6 Terra was 21.5%.

The comparison points to stronger performance on the evaluated coding challenges.

Which OpenAI Model Is Better for Complex Coding?

Despite the reported gains, OpenAI cautioned that GPT-6.1 Sol and GPT-6 Astra remain preferable options for the most demanding coding assignments.

This distinction suggests that GPT-6.1 Terra is positioned primarily for efficient everyday work rather than replacing every specialised model in the company's lineup.

Users choosing between models may need to consider the complexity of their projects, response speed, cost and the level of reasoning required.

GPT-6.1 Terra Availability on ChatGPT and API

GPT-6.1 Terra is available to Plus, Pro, Enterprise and Edu subscribers through ChatGPT Work and Codex, according to the supplied announcement.

Developers can also access the model through the OpenAI API using the model identifier gpt-6.1-terra.

This API access allows developers to integrate the model into applications, automated workflows and other software tools.

GPT-6.1 Terra API Pricing

OpenAI has announced the following API prices for GPT-6.1 Terra:

• Input tokens: $1 per million tokens

• Cached input tokens: $0.05 per million tokens

• Output tokens: $5 per million tokens

The company says standard API rates are 50% lower than those for GPT-5.6 Terra and GPT-6.1 Sol.

Lower pricing could make it more economical for businesses and developers to use AI for repeated tasks, particularly where large volumes of text or frequent automated operations are involved.

Conclusion

GPT-6.1 Terra is presented as an efficiency-focused AI model for everyday professional work, combining visual understanding, computer interaction, coding improvements and lower API costs. OpenAI's reported benchmark results indicate progress in speed and accuracy, although demanding coding tasks may still benefit from other models. Its practical value will depend on how well it performs across users' specific workflows.