Models & pricing / gpt-6.1-terra · OpenAI
OpenAI

gpt-6.1-terra API

GPT-6.1 Terra is OpenAI’s model for visual reasoning, computer-use tasks and natural writing, with text and image input and text output. Evaluate image understanding, parameters and current costs through XiuAI; tool calling requires Responses.[1][3]

gpt-6.1-terra

Model overview

Published provider specifications. Available features depend on the current API route.

API routes
OpenAI-compatible API / openai-response / openai-response-compact / Anthropic Messages / Gemini GenerateContent / openai-alpha-search

Turn visual material into reviewable text

OpenAI’s examples include matching product photos to a catalog, interpreting technical drawings, working with application screens and drafting or revising text. Supply the candidate catalog, legend or task goal and verify the interpretation. Understanding a screenshot does not mean an action has been executed; screen interaction still needs your tools and permissions.[1]

Compare task requirements with Sol

Terra’s positioning emphasizes visual reasoning and writing, while GPT-6.1 Sol targets complex coding and professional work. Both require Responses for tools, but Terra includes a none reasoning setting and Sol does not. Choose from task and parameter requirements, then compare current XiuAI rates and equivalent outputs rather than adopting another platform’s fixed speed claim.[1][2][3]

Distinguish cache reads from writes

The current GPT-6.1 Terra catalog lists input, output, cache and long-context conditions. Cache reads, initial writes and retention options are different billing items; a read rate does not apply to all repeated material. Use applicable prices and actual usage, without assuming that vendor retention or processing-mode pricing establishes the same XiuAI configuration.[1][3]

API access

Using GPT-6.1 Terra through XiuAI

Use gpt-6.1-terra with an existing authorized API key. Format text or image-understanding input for the selected protocol. Use Responses for function tools and handle its returned tool calls. Regardless of reasoning effort, do not treat a Chat Completions tools configuration as the tool integration for this version.[1][3]

Send a request using your existing API key.

POST /v1/chat/completions
curl 'https://router-api.xiu.ai/v1/chat/completions' \
  -H "Authorization: Bearer $XIUAI_API_KEY" \
  -H 'Content-Type: application/json' \
  --data-raw '{
  "model": "gpt-6.1-terra",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ]
}'
Read the API documentation

Frequently asked questions

Does setting reasoning to none enable tools in Chat Completions?

Do not infer that. GPT-6.1 Terra still requires Responses for tools. Supporting none reasoning and supporting a tool protocol are independent conditions; do not carry over the similarly named GPT-6 rule.[1]

Does it output images, audio or video?

Its direct output is text. Image input does not imply image generation, and the vendor explicitly excludes audio and video input. Use an appropriate available model for media generation or audio/video understanding instead of adding another model’s fields to Terra.[1][3]

Can product matching or drawing interpretation be used as the final business decision?

These are vendor-listed task directions, not accuracy guarantees. Supply candidates, legends and decision rules, retain the model’s evidence and validate or review results when errors are costly. XiuAI access does not replace your business approval process.[1]

Related models

gpt-6.1-solgpt-6-lunagpt-5.6-terra

Sources and updates

Specifications reference provider documentation; prices and availability come from the current XiuAI catalog.

Content reviewed: 2026-11-16

[1] OpenAI GPT-6.1 Terra documentation[2] OpenAI GPT-6.1 Sol documentation[3] Current XiuAI model and pricing dataModel documentationTool callingStructured outputContext documentation