Terra model gets GPT 6.1 as OpenAI upgrades ChatGPT and Codex with faster, cheaper performance - 9to5Mac
A new model for work
GPT-6.1 Terra is OpenAI’s medium-sized model, designed to balance capability and price. The company describes it as an upgrade from GPT-5.6 Terra. OpenAI says the model is suited to everyday communication and computer-based tasks. Examples include drafting proposals, interpreting technical drawings, completing forms and editing videos.
Lower cost and fewer errors
OpenAI says GPT-6.1 Terra costs 50% less than GPT-5.6 Terra. The company presents the lower price as a way to automate computer tasks for less. On an internal benchmark using prompts designed to induce errors, the rate at low reasoning effort fell from 53.6% to 16.7%. At maximum effort, it dropped by 18 percentage points to 4.3%, which OpenAI says nearly matches GPT-6 Astra and GPT-6.1 Sol.
Part of OpenAI’s lineup
OpenAI lists four main model classes, ordered from lower price and capability to higher: Luna, Terra, Sol and Astra. Terra occupies the medium-sized position in that lineup. The company says GPT-6.1 Terra combines visual reasoning and precise computer use with more natural writing. It also highlights speed and efficiency.
Available across products
GPT-6.1 Terra is available now through ChatGPT Work, Codex and the API. The release follows OpenAI’s earlier GPT-6 models, Sol and Luna, while Terra was not included in that update. The upgrade gives users access to the model across OpenAI’s listed products, with the company emphasizing its lower price and reported improvements in accuracy.
Key points
- OpenAI upgraded Terra from GPT-5.6 to GPT-6.1.
- The company says the new model costs 50% less than its predecessor.
- At maximum reasoning effort, the reported benchmark error rate was 4.3%.
- GPT-6.1 Terra is available through ChatGPT Work, Codex and the API.