openai / gpt-5-6-terra

Overview

GPT 5.6 Terra is an OpenAI model in the GPT 5.6 family. It reads your prompt and, when provided, your images. It then generates a text response that explains what it sees and answers your request. It’s useful when you need reliable image-aware explanations without stepping up to the flagship tier.

What you can build

• Image Q&A for products, screenshots, or diagrams
• Customer support helpers that interpret user-submitted screenshots
• Visual content review and labeling with written rationales
• Document-like image analysis, such as forms or slides saved as images
• Coding help that uses screenshots of errors, UI, or console output

Inputs

• A required written prompt that tells the model what to do and what format you want back.
• Optional image uploads to ground the answer in visual evidence. You can provide multiple images in one request, up to 10.
• Optional system-level instructions that set global rules for tone, constraints, and task focus.
• Optional conversation identifiers to keep chat history organized across turns. Provide a user identifier and a session identifier when you need continuity.
• Optional reasoning intensity control that trades speed for deeper multi-step thinking.
• Optional web search toggle when you want the model to look things up during the run.
• Optional verbosity control that nudges the model toward shorter or more detailed answers.

Outputs

• A text response that answers your prompt.
• If you include images, the response typically includes image-grounded descriptions, extracted details, and conclusions tied to visible elements.
• If safety checks block part of a request, the output may be a refusal or a partial answer with constraints.

Limitations

• This Wiro deployment is image-focused. If your task needs PDFs, spreadsheets, or audio, convert them to images first.
• Low-quality inputs reduce accuracy. Blurry screenshots, heavy compression, glare, tiny text, or cropped diagrams can cause missed details.
• Some cybersecurity and biological requests can trigger extra automated checks. The platform may delay, limit, or block those responses.

Safety & compliance

• Follow OpenAI usage rules for disallowed content and sensitive requests.
• For cybersecurity tasks, frame prompts around defensive goals like detection, mitigation, and remediation.
• For biological topics, focus on safety, prevention, analysis, and risk reduction rather than procedural enablement.
• Don’t submit secrets, credentials, or sensitive personal data in prompts or images unless you have permission and a clear need.