GPT-5.6 Terra · OpenAI

What does GPT-5.6 Terra cost to run in a coding agent?

GPT-5.6 Terra costs about $0.084 per average coding-agent request on our measured traffic, at list prices of $2 per million input and $12 per million output tokens. It scores 42 on the Artificial Analysis Intelligence Index, 15th of the 48 models we track. Source: Artificial Analysis (artificialanalysis.ai)

List prices: OpenAI API pricing, checked 2026-10-13. Request costs: Standard Compute gateway data, 79,301 coding-agent requests from 75 users, Sep 13 to Oct 12, 2026. How we compute this

Run GPT-5.6 Terra in your agent from $19/month

See plans

What GPT-5.6 Terra costs per coding-agent request

Cost at list price	GPT-5.6 Terra
Prompt under 100K tokens50% of requests, 87% of input cached	$0.032
Prompt of 100K to 272K39% of requests, 88% of input cached	$0.080
Prompt over 272K tokens11% of requests, 86% of input cached	$0.34
Average request	$0.084
100 requests	$8.40
Median session16 requests; a session ends after 10 idle minutes	$0.92
A month of regular usemedian of 20 users active on 5+ days	$118

Each logged request priced at the vendor’s list rates for its own token counts, including cache reads and any higher rate for long prompts. Average output: 698 tokens per request.

GPT-5.6 Terra benchmarks, speed and latency

Artificial Analysis	Value	Rank
Intelligence Index10 evaluations, 0 to 100	42	15 of 48
Terminal-Bench 4.0agentic coding in a terminal	35.4%	12 of 39
Output speedmedian	135 tokens/s	11 of 44
Time to first answer tokenincludes thinking time	2.2 min	35 of 44

Source: Artificial Analysis (artificialanalysis.ai). Measured variant: GPT-5.6 Terra (Max), speed on OpenAI. Fetched October 13, 2026. Rank is among the models on this site, older versions included. Artificial Analysis runs Terminal-Bench 4.0 in its own setup, so its scores differ from the tbench.ai leaderboard.

About GPT-5.6 Terra

GPT-5.6 Terra is the everyday tier of OpenAI's GPT-5.6 family, released July 9, 2026 for high-volume business tasks. OpenAI positioned it as competitive with GPT-5.5 at about half the price, and Artificial Analysis bears that out: 42 on its Intelligence Index against 38 for GPT-5.5, with 35.4% in its Terminal-Bench 4.0 run and about 135 output tokens per second. The GPT-6 family has since overtaken it: GPT-6.1 Sol scores 52 and costs less per request in our measurements ($0.067 against $0.084 for Terra), mainly because its cached input is half the price.

Sources: Artificial Analysis

GPT-5.6 Terra FAQ

How much does GPT-5.6 Terra cost per coding-agent request?

$0.084 on average at list price, $8.40 per 100 requests, and $0.92 for a median 16-request session. That is 79,301 real coding-agent requests (Sep 13 to Oct 12, 2026) priced at OpenAI's list rates; the median request sent 99,534 input tokens, 87% of input was read from cache, and requests returned 698 output tokens on average.

Does GPT-5.6 Terra charge more for long prompts?

Yes. Prompts over 272K tokens are billed at $4 per million input and $18 per million output tokens, on every token in the request including cache reads. 11% of the coding-agent requests we measured were that large.

How fast is GPT-5.6 Terra?

Artificial Analysis measures GPT-5.6 Terra (Max) at 135 output tokens per second, with 129 seconds to the first answer token, thinking time included, on OpenAI's API. Source: Artificial Analysis (artificialanalysis.ai).

Is GPT-5.6 Terra good for AI agents?

Fast mid-tier work in existing GPT-5.6 setups. On our scoring, which is calibrated to public benchmarks, it rates 7.8/10 for agentic ability and 8.0/10 for output quality. 1M-token context at $2/$12 per million.

What is GPT-5.6 Terra's context window?

GPT-5.6 Terra has a 1.1M-token context window, large enough to hold sizeable codebases or document sets in a single run. OpenAI is the maker.

Can I self-host GPT-5.6 Terra?

No. GPT-5.6 Terra is a proprietary, hosted-only model from OpenAI. You access it through an API, such as OpenRouter, rather than running the weights yourself.

What are the downsides of GPT-5.6 Terra?

The main trade-offs: GPT-6.1 Sol scores higher and costs less per request ($0.067 vs $0.084); Prompts over 272K tokens bill at higher rates. It's strongest for fast mid-tier work in existing GPT-5.6 setups.