How is GPT-6.1 Terra different from GPT-5.6 Terra?

Context window (1.05M), max output (128K), input modalities and reasoning effort values (none to max) are the same. The difference is generation and knowledge: 5.6 Terra is the July 2026 GPT-5.6 member with knowledge to February 16, 2026, while 6.1 Terra is an October 26 GPT-6 model with knowledge to June 1, 2026. OpenAI documents reasoning reuse among the three GPT-5.6 tiers, so stay on 5.6 Terra if one conversation mixes it with 5.6 Sol or Luna and should keep its reasoning. When moving, OpenAI suggests comparing the two on representative images and workflows at the same reasoning effort.

When should I move up to GPT-6.1 Sol?

OpenAI labels 6.1 Sol as the balance of speed, cost and intelligence with near-Astra performance, and Terra as fast and efficient. Both share the context window and output limit, so switching is a change to the model field. One catch: 6.1 Sol rejects reasoning_effort: none, so requests that send none to Terra need low on Sol. Run the same batch on each to see where Terra falls short.

Does the price change for very long inputs?

Yes. A request whose input exceeds 272K tokens is billed entirely at the long-context rate, not just the part above the threshold. The pricing section above lists both tiers.

GPT-6.1 Terra

Save 85%Official

openai/gpt-6.1-terra

The Terra model of OpenAI's GPT-6 generation, positioned as fast and efficient for visual reasoning, precise computer use, and clear, natural writing.

Get an API key View code sample

Live model playground

Try GPT-6.1 Terra here

Tune the request, see the current price ceiling, and inspect the real response without leaving this model page.

streamchat

1Configure input

Prompt

0 / 20,000

Add images

Maximum estimated cost—

2Live result

Your result will appear here

Enter an input on the left. The same RouteMux Playground execution and billing path powers this embedded test.

Context window: 1.05M
Max output: 128K
Input modalities: TextImage
Output modalities: Text
Capabilities: Vision · Reasoning · Tool calling · Coding · Prompt cache · Paid cache writes
Upstream release: 2026-10-26
RouteMux name: openai/gpt-6.1-terra

What is GPT-6.1 Terra

GPT-6.1 Terra was released by OpenAI on October 26, 2026. OpenAI's GPT-6 model guide features four models: Astra, 6.1 Sol, 6.1 Terra and GPT-6 Luna. Terra is labeled fast and efficient, aimed at visual reasoning, precise computer use and clear, natural writing; OpenAI's examples include matching product photos to catalog entries, comparing technical drawings and acting on application screens.

The specs: a 1.05M-token context window (up to 922K input), 128K max output, text and image in, text out, and a knowledge cutoff of June 1, 2026. Reasoning effort takes none, low, medium, high, xhigh or max, with medium as the default; unlike Astra and 6.1 Sol it accepts none, so reasoning can be switched off. On RouteMux it supports tool calling and explicit prompt caching, and it is reachable over OpenAI Chat Completions, OpenAI Responses and Anthropic Messages.

When to stay on GPT-5.6 Terra: one conversation moves between GPT-5.6 Sol, Terra and Luna and should keep its reasoning (OpenAI's reasoning guide documents reuse across the three 5.6 tiers and does not list 6.1 Terra), or your prompts and evals were tuned against 5.6 Terra. For the move itself, OpenAI suggests running both models on representative images and workflows at the same reasoning effort.

At the standard tier (input up to 272K tokens) it bills $0.15 per 1M input tokens and $0.75 per 1M output tokens, with cached input at $0.0075 per 1M tokens. OpenAI lists $1, $5 and $0.05 for the same three.

Before you choose

How is GPT-6.1 Terra different from GPT-5.6 Terra?: Context window (1.05M), max output (128K), input modalities and reasoning effort values (none to max) are the same. The difference is generation and knowledge: 5.6 Terra is the July 2026 GPT-5.6 member with knowledge to February 16, 2026, while 6.1 Terra is an October 26 GPT-6 model with knowledge to June 1, 2026. OpenAI documents reasoning reuse among the three GPT-5.6 tiers, so stay on 5.6 Terra if one conversation mixes it with 5.6 Sol or Luna and should keep its reasoning. When moving, OpenAI suggests comparing the two on representative images and workflows at the same reasoning effort.
When should I move up to GPT-6.1 Sol?: OpenAI labels 6.1 Sol as the balance of speed, cost and intelligence with near-Astra performance, and Terra as fast and efficient. Both share the context window and output limit, so switching is a change to the model field. One catch: 6.1 Sol rejects reasoning_effort: none, so requests that send none to Terra need low on Sol. Run the same batch on each to see where Terra falls short.
Does the price change for very long inputs?: Yes. A request whose input exceeds 272K tokens is billed entirely at the long-context rate, not just the part above the threshold. The pricing section above lists both tiers.

GPT-6.1 Terra API pricing

Prices verified 2026-10-27

InputOfficial $1 / M tokens

$0.15/ M tokensSave 85%

OutputOfficial $5 / M tokens

$0.75/ M tokensSave 85%

Fast $0.3 / $1.5 /M

Cache read $0.0075/M · Cache write $0.1875/M

Tier	Input	Cache read	Cache write	Output
Standard ≤272K	$0.15$1−85%	$0.0075~~$0.05~~−85%	$0.1875~~$1.25~~−85%	$0.75$5−85%
Standard >272K	$0.3$2−85%	$0.015~~$0.1~~−85%	$0.375~~$2.5~~−85%	$1.125~~$7.5~~−85%
Priority ≤272K	$0.3$2−85%	$0.015~~$0.1~~−85%	$0.375~~$2.5~~−85%	$1.5~~$10~~−85%
Priority >272K	$0.6$4−85%	$0.03~~$0.2~~−85%	$0.75$5−85%	$2.25~~$15~~−85%

Estimate cost

Input tokensOutput tokens

Estimated cost$0.00225

Key capabilities

Built for visual work and computer use: OpenAI positions it for visual reasoning, precise computer use and clear, natural writing, with examples like matching product photos to catalog entries.
Reasoning can be turned off: reasoning_effort takes none through max, defaulting to medium. GPT-6 Astra and 6.1 Sol reject none; 6.1 Terra accepts it.
Knowledge cutoff June 1, 2026: About three and a half months later than GPT-5.6 Terra's February 16; the 1.05M context and 128K output match 5.6 Terra.

Limited time10% extra Credits on your first top-up — up to 200 bonus Credits

Applies to your first top-up within 72 hours of signing up. Minimum $20, once per account.

Interface and protocols

OpenAI Chat CompletionsStreamingOpenAI ResponsesStreamingAnthropic MessagesStreaming

Use any of these IDs to call this model via the API.

RouteMux nameopenai/gpt-6.1-terra

Replace the ROUTEMUX_KEY placeholder with your API key. Create one →

from openai import OpenAI

client = OpenAI(
    base_url="https://api.routemux.com/v1",
    api_key="<ROUTEMUX_KEY>",
)

completion = client.chat.completions.create(
    model="openai/gpt-6.1-terra",
    messages=[{"role": "user", "content": "What is the meaning of life?"}],
)
print(completion.choices[0].message.content)

base_url: https://api.routemux.com/v1

Performance (last 30 days)

40 requests · last 30 days

Total usage1.16M

P50 first-token latencyInsufficient data (40 requests)

Avg throughputInsufficient data (40 requests)

Success rate100.0%

Uptime · 24htested via openai_chat

100.00%

Success rate · 60d · from real requests

GPT-6.1 Terra FAQ

›What is GPT-6.1 Terra?

The Terra model of OpenAI's GPT-6 generation, positioned as fast and efficient for visual reasoning, precise computer use, and clear, natural writing.

›How large is the context window of GPT-6.1 Terra?

GPT-6.1 Terra supports a context window of up to 1.05M tokens, with up to 128K output tokens per request.

›How much does the GPT-6.1 Terra API cost?

Through RouteMux, GPT-6.1 Terra costs $0.15 for input and $0.75 for output (USD per 1M tokens), below the official list price.

›How do I call GPT-6.1 Terra via API?

GPT-6.1 Terra is OpenAI-compatible: point your base URL at RouteMux and set the model field to openai/gpt-6.1-terra — no code changes needed.

›Which input and output modalities does GPT-6.1 Terra support?

GPT-6.1 Terra accepts text, image as input and produces text as output.

›When was GPT-6.1 Terra released?

GPT-6.1 Terra was released on October 26, 2026.

Sources

View as Markdown