GPT-6.1 Terra API: Pricing, Batch and How to Call It
GPT-6.1 Terra API pricing checked against OpenAI: $1/$5 per 1M tokens, the 272K long-context rule, 50% batch, 1.05M context, and code to call it.

The GPT-6.1 Terra API went live on October 26, 2026. OpenAI announced GPT-6.1 Terra as a fast model for visual reasoning, computer use and writing, with standard API prices 50% lower than GPT-6.1 Sol. It is available in the OpenAI API as gpt-6.1-terra, and the same day on OpenRouter, together with Terra Pro and discounted batch variants. This page collects the numbers you need before wiring it into an app or an agent.
Sources, checked October 29, 2026: OpenAI's GPT-6.1 Terra model page and OpenRouter's public model list (
/api/v1/models). The standard prices match on both. Terra Pro is an OpenRouter listing; we did not find it on OpenAI's model page.
GPT-6.1 Terra at a glance
| GPT-6.1 Terra | GPT-6.1 Terra Pro | |
|---|---|---|
| OpenAI API model ID | gpt-6.1-terra |
not listed by OpenAI |
| OpenRouter ID | openai/gpt-6.1-terra |
openai/gpt-6.1-terra-pro |
| Listed on OpenRouter | Oct 26, 2026 | Oct 26, 2026 |
| Context window | 1,050,000 tokens | 1,050,000 tokens |
| Max output | 128,000 tokens | 128,000 tokens |
| Input types | text, image (OpenAI); text, image, file (OpenRouter) | text, image, file (OpenRouter) |
| Tools / structured outputs | yes | yes |
OpenAI also lists a maximum of 922,000 input tokens and a June 1, 2026 knowledge cutoff.
Per OpenRouter, Terra Pro is the same underlying model, served with reasoning.mode set to pro. OpenRouter warns that pro mode spends far more reasoning tokens, so the per-token price is the same but the bill per task is usually higher.
GPT-6.1 Terra API pricing (per 1M tokens)
| Variant | Input | Output | Cached input read | Cache write |
|---|---|---|---|---|
| Standard (prompt ≤ 272K) | $1.00 | $5.00 | $0.05 | $1.25 |
| Standard (prompt > 272K) | $2.00 | $7.50 | $0.10 | $2.50 |
| Batch (prompt ≤ 272K) | $0.50 | $2.50 | $0.025 | $0.625 |
| Batch (prompt > 272K) | $1.00 | $3.75 | $0.05 | $1.25 |
The same prices apply to Terra and Terra Pro on OpenRouter, and both have :batch variants. OpenAI's own page lists the same standard rates, plus: cache writes cost $1.25 for 30-minute retention and $2.50 for 12-hour retention; Batch and Flex are 50% lower; Fast mode is 2× Standard; regional processing adds 10%. Web search through OpenRouter is billed separately at $0.01 per request.
How much does GPT-6.1 Terra batch save?
Batch is exactly half price on every line. Example: a nightly job with 2,000 requests × 20K input + 2K output tokens:
- Standard: 40M × $1 + 4M × $5 = $60
- Batch: 40M × $0.50 + 4M × $2.50 = $30
The catch: batch is for work that can wait. It is a poor fit for interactive chat or a coding agent that needs an answer in seconds. Use it for evals, bulk classification, translation and document processing.
Watch the 272K line. OpenAI says that above 272K input tokens, input and cache rates double and output rises 1.5×, for the whole request. Using the full 1M context on every call is the quickest way to a surprise bill; cache or trim long prompts.
How to call the GPT-6.1 Terra API
Directly from OpenAI, use the model ID gpt-6.1-terra. It supports Chat Completions, Responses and Batch, but OpenAI notes that tool calling requires the Responses API:
from openai import OpenAI
client = OpenAI() # reads OPENAI_API_KEY
r = client.responses.create(model="gpt-6.1-terra", input="Summarize this changelog in 5 bullets.")
print(r.output_text)
Through OpenRouter, use its OpenAI-compatible endpoint:
from openai import OpenAI
client = OpenAI(base_url="https://openrouter.ai/api/v1", api_key="YOUR_OPENROUTER_KEY")
r = client.chat.completions.create(
model="openai/gpt-6.1-terra", # or openai/gpt-6.1-terra-pro
messages=[{"role": "user", "content": "Summarize this changelog in 5 bullets."}],
)
print(r.choices[0].message.content)
For the batch variant, use the :batch model ID (openai/gpt-6.1-terra:batch). Check OpenRouter's docs for how batch requests are submitted and how long results can take; we did not test it for this post.
When the GPT-6.1 Terra API is overkill: agent loops on a flat budget
Coding agents (Claude Code, Codex, tstack's babysit-pr) re-send big contexts dozens of times an hour. Even at $1/$5, a day of agent loops can cost more than a month of a flat plan.
XCoffee plan details updated November 12, 2026.
XCoffee is a different trade-off:
- One key, three groups:
xcoffee-coding,xcoffee-fast,xcoffee. You pick the group; the relay picks the model inside it, with automatic failover. You cannot pick a specific model, and GPT-6.1 Terra is not in any XCoffee group today. - Plan-based pricing: 30-day passes have different prices, daily credit allowances and rolling hourly caps. See the current plans; daily credits reset at UTC 00:00.
- Compatible: OpenAI (
https://api.xcoff.ee/v1), Anthropic (https://api.xcoff.ee) and Gemini formats. - Limits: concurrency and request-rate caps depend on your plan. Your account shows the applicable limits. Provider-hosted tools, including built-in web search, are not included.
A practical split: keep Terra (or Terra batch) for the hard or bulk jobs that really need it, and run the high-volume agent loops through a flat key. Setup for Claude Code and Codex is in our tstack guide; free options are in Free LLM API options; full config is in the guide. Comparing open-weight options? See Kimi K3.1 Pro API.
GPT-6.1 Terra API FAQ
What is GPT-6.1 Terra?
OpenAI's fast, efficient GPT-6.1 model for visual reasoning, computer use and writing, released October 26, 2026. Its standard API prices are 50% lower than GPT-6.1 Sol's.
What is the GPT-6.1 Terra API model name?
gpt-6.1-terra in the OpenAI API, and openai/gpt-6.1-terra on OpenRouter.
How much does the GPT-6.1 Terra API cost?
$1 input / $5 output per 1M tokens ($0.05 cached input) for prompts up to 272K, and $2 / $7.50 above that, on both OpenAI and OpenRouter.
What is the difference between Terra and Terra Pro?
On OpenRouter they are the same model at the same per-token price; Pro runs with reasoning.mode: pro, which uses many more reasoning tokens.
How much cheaper is GPT-6.1 Terra batch?
50% off standard prices, on OpenAI (Batch and Flex) and on OpenRouter's :batch IDs. Use it for jobs that don't need an immediate answer.
What is the GPT-6.1 Terra context window?
1,050,000 tokens, with up to 922,000 input and 128,000 output tokens.
Does GPT-6.1 Terra support images and tools?
Yes. OpenAI lists text and image input (OpenRouter also accepts files). Tool calling is supported, and in the OpenAI API it requires the Responses API.
Can I use GPT-6.1 Terra through XCoffee?
No. XCoffee routes to models inside its groups and does not offer GPT-6.1 Terra or model selection.
Is there a free GPT-6.1 Terra API?
Not that we found. See free LLM API options for models that are free today.