Reading Images and Technical Drawings
Interpreting and comparing technical drawings, and matching product photos to catalog entries.
gpt-6-1-terra
Automatically append previous messages to maintain multi-turn context. May increase token usage.
Function calling
Web Access
Reasoning effort
Stream Response
How can I help you today?
README
Complete guide to using gpt-6-1-terra
GPT-6.1 Terra API for Visual Reasoning, Writing and Tool Calling
OpenAI describes GPT-6.1 Terra as a fast, efficient model for detailed visual reasoning, precise computer use and clearer, more natural writing, and a significant step up from GPT-5.6 Terra. OpenAI lists a 1,050,000-token context window. On Kie.ai, the GPT-6.1 Terra API supports web search and function calling, and you can call it alongside GPT-6.1 Sol and GPT-6 Astra with one API key.
New to GPT-6.1 Terra? Read what it is and how it compares →
New to GPT-6.1 Terra? Read what it is and how it compares →
curl https://api.kie.ai/codex/v1/responses \-H "Authorization: Bearer $KIE_API_KEY" \-H "Content-Type: application/json" \-d '{"model": "gpt-6-1-terra","input": "Draft a short release note for a new image search feature.","reasoning": { "effort": "medium" },"stream": false}'
Reasoning
GPT-6.1 Terra API Parameters on Kie.ai
| Parameter | Value |
|---|---|
| Model | gpt-6-1-terra (responses report the model as gpt-6.1-terra) |
| Endpoint | POST /codex/v1/responses (OpenAI Responses API format) |
| Context window | 1,050,000 tokens (OpenAI specification); an 898,828-token input worked in our tests on Kie.ai |
| Maximum input | 922,000 tokens (OpenAI specification) |
| Maximum output | 128,000 tokens (OpenAI specification) |
| Input types | Text, images and files, mixed in one message |
| Reasoning effort | low, medium, high, xhigh or max; default medium |
| Tools | Web search or function calling, not both in one request |
| Streaming | On by default (server-sent events); set stream to false for a single JSON response |
| Knowledge cutoff | June 1, 2026 (OpenAI specification) |
| Pricing | Shown in the pricing card above |
Choosing Between GPT-6.1 Terra, GPT-6.1 Sol and GPT-6 Astra
OpenAI positions GPT-6.1 Terra as its fast, efficient GPT-6.1 model for visual reasoning, computer use and everyday writing. For the most complex coding tasks, OpenAI says GPT-6.1 Sol and GPT-6 Astra remain the better choice. On the MMMU-Pro visual-reasoning benchmark, OpenAI reports that Terra nearly matches both. Terra also supports the "none" reasoning effort setting, which Sol and Astra do not. Check each model's pricing card on Kie.ai for billing details.
| GPT-6.1 Terra | GPT-6.1 Sol | GPT-6 Astra | |
|---|---|---|---|
| How OpenAI positions it | A fast, efficient model for visual reasoning, precise computer use and clear, natural writing | Nearly matches Astra on agentic coding, computer use and professional work | Its best model across the board, for when you want the best results |
| OpenAI's pick for the most complex coding tasks | Not the first choice | Recommended | Recommended |
| "none" reasoning effort (OpenAI specification) | Supported | Not supported | Not supported |
| Available on Kie.ai | Yes | Yes | Yes |
What GPT-6.1 Terra Improves Over GPT-5.6 Terra
OpenAI built GPT-6.1 Terra as a significant step up from GPT-5.6 Terra for communication and computer-based work. Below are the benchmark results OpenAI reports against GPT-5.6 Terra.
| Benchmark (OpenAI-reported) | What it measures | GPT-6.1 Terra vs GPT-5.6 Terra |
|---|---|---|
| Agents' Last Exam | Professional computer-use tasks, such as turning architectural drawings into a 3D model | 15 percentage points higher at medium reasoning effort |
| MMMU-Pro | Visual reasoning | Higher at every reasoning effort setting |
| OSWorld 2.1 | Multistep tasks in desktop apps | At medium reasoning effort, surpasses GPT-5.6 Terra's best score |
| Terminal-Bench 4.0 | Coding agents solving problems from the command line | 50.3% of tasks at maximum reasoning effort vs GPT-5.6 Terra's published 21.5% |
| Internal error-rate evaluation | Error rate on OpenAI's internal set of error-inducing prompts | Down from GPT-5.6 Terra's 53.6% to 16.7% at low reasoning effort; down 18 percentage points to 4.3% at maximum effort |
Typical Uses for the GPT-6.1 Terra API
OpenAI built GPT-6.1 Terra for the communication and computer-based tasks that fill a workday. OpenAI's examples include:
Acting on Application Screens
Reading application screens and deciding the next step, such as filling out forms. OpenAI highlights this computer-use skill; on Kie.ai, function calling and web search are the tools we have tested.
High-Volume Data Entry
Reading incoming invoices and entering their details into accounting software.
Drafting and Revising Text
Drafting and revising proposals and other everyday business writing with clearer, more natural phrasing.
How to Use the GPT-6.1 Terra API on Kie.ai
- 1
Step 1. Get Your API Key
Sign up on Kie.ai, open the API Key Management page and create an API key. Send it on every request in the header Authorization: Bearer YOUR_API_KEY. - 2
Step 2. Test Your Prompt in the Playground
Test the model in the playground on this page before writing code. Start with medium reasoning effort and increase it if the improvement in quality is worth the added time and tokens. - 3
Step 3. Send a Request
Send a POST request to /codex/v1/responses with model set to gpt-6-1-terra and your input, using the parameters in the table above. Set stream to false if you want a single JSON response instead of a server-sent event stream.
Tips Before You Ship
- 01
Check the Body, Not Only the Status
Errors can come back with HTTP status 200 and an error code in the response body, so check the body before treating a call as successful. - 02
Streaming Is the Default
A request without a stream field returns server-sent events. If your code expects a single JSON object, set stream to false. - 03
Use Publicly Accessible Image URLs
Image inputs are fetched directly from the URL you provide, so use a publicly accessible address. If an image host blocks automated fetches, the request will fail.
Why Use the GPT-6.1 Terra API on Kie.ai
Same Key Across the Kie.ai Market
All models in the Kie.ai market use the same authentication method: an Authorization: Bearer YOUR_API_KEY header with the key from your Kie.ai API Key Management page.
Free Credits to Start
New Kie.ai users receive 80 free credits for testing. Your account has one credit balance that all generation services draw from, and you can check it with the Get Remaining Credits API.
Nothing to Host or Install
Kie.ai provides this model as an API with a playground; there is no app to download or install. There are no GPUs or model weights to manage.
Frequently Asked Questions
What is the context window of the GPT-6.1 Terra API on Kie.ai?
OpenAI lists a 1,050,000-token context window, with up to 922,000 input tokens and 128,000 output tokens. In our tests on Kie.ai, a request with an 898,828-token input completed normally.
How much does the GPT-6.1 Terra API cost?
Pricing is shown in the pricing card above.
Is the GPT-6.1 Terra API free?
No. Each request on Kie.ai uses credits according to the pricing card above. New Kie.ai users receive 80 free credits for testing.
What reasoning effort levels does GPT-6.1 Terra support?
On Kie.ai, the documented levels are low, medium, high, xhigh and max, with medium as the default. OpenAI also lists a none level; it worked in our tests on Kie.ai on October 27, 2026, but it is not in the Kie.ai documentation.
Does GPT-6.1 Terra support image input?
Yes. You can pass both images and text in the same request to the /codex/v1/responses endpoint. Use publicly accessible image URLs, because the request fails if the image host blocks the fetch.
Does GPT-6.1 Terra support function calling?
Yes. On Kie.ai it supports custom function tools or web search through the Responses-compatible /codex/v1/responses endpoint. Choose only one of the two in a single request.
Is the GPT-6.1 Terra API good for coding?
OpenAI reports that it solves 50.3% of tasks on Terminal-Bench 4.0 at maximum reasoning effort, compared with GPT-5.6 Terra's published 21.5%. For the most complex coding tasks, OpenAI still recommends GPT-6.1 Sol or GPT-6 Astra.
How do I get a GPT-6.1 Terra API key?
Sign up on Kie.ai and create a key on the API Key Management page. The same key works across every model in the Kie.ai market.