OpenAI GPT-6.1 Terra Pro lists on OpenRouter at $1 input, $5 output per 1M tokens, 1.1M context

OpenAI's GPT-6.1 Terra Pro, listed on OpenRouter as openai/gpt-6.1-terra-pro, is the same underlying model as GPT-6.1 Terra, served with reasoning.mode set to pro for higher-quality responses on complex tasks. OpenRouter lists a 1.1M token context window, text modality, and a release date of Oct 26, 2026.

Pricing

OpenRouter lists the following rates per 1M tokens for the OpenAI endpoint:

Tier	Input	Output	Cache read
Standard	$1.00	$5.00	$0.050
Flex	$0.50	$2.50	$0.025
Fast	$2.00	$10.00	$0.100

OpenRouter notes that the Flex and Fast endpoints are not used in Standard routing. A separate batch listing, GPT-6.1 Terra Pro (batch), is priced at $0.50 input and $2.50 output per 1M tokens.

The per-token rate matches the $1 input and $5 output listed for GPT-6.1 Terra. The cost difference comes from token volume, not rate. According to the listing, pro mode spends far more reasoning tokens per request, so a typical request costs several times more than the same request on GPT-6.1 Terra and takes much longer to complete.

Intended use

The listing says Terra Pro is intended for hard, high-stakes problems where the extra accuracy justifies the cost. For everyday coding, agentic, and chat workloads, it points users to GPT-6.1 Terra instead. It links to OpenAI's reasoning-mode documentation for details on the setting.

OpenRouter describes the base model, GPT-6.1 Terra, as the mid-tier model in OpenAI's GPT-6.1 series, priced below GPT-6.1 Sol. It is suited for everyday agentic, coding, and writing work, is particularly strong at understanding screenshots and on-screen content, and is efficient enough for high-volume workloads.

Performance as listed on OpenRouter

As read in October 2026, OpenRouter showed a P50 latency of 4.10s and P50 throughput of 61 tokens per second, both labelled best across providers, with OpenAI as the only provider shown. Uptime was not displayed: the page said there was not enough uptime data yet over 3 days. These are live figures and will change.

The listing also states that the model is not currently available through EU in-region routing.

Related listings

The same page lists pro-mode variants of other OpenAI models, built the same way (same underlying model, reasoning.mode set to pro):

• GPT-6.1 Sol Pro: $2 input, $10 output per 1M tokens, 1.1M context. Its batch version is listed at $1 and $5.
• GPT-6 Luna Pro: $0.10 input, $0.50 output per 1M tokens, 1.1M context. Its batch version is listed at $0.05 and $0.25.

GPT-6.1 Sol is described as an upgrade to GPT-6 Sol, positioned below the flagship GPT-6 Astra, and as making fewer factual errors and respecting explicit restrictions and user intent more reliably during agentic tasks than GPT-6 Sol.

No benchmark scores, parameter counts, or training cutoff date appear on the listing.

What this means

Terra Pro is a serving mode, not a new model. Developers who already use GPT-6.1 Terra can switch the reasoning mode setting to pro without changing the model's listed per-token price. The cost to plan for is reasoning-token consumption, which OpenRouter says makes a typical request several times more expensive and much slower than the same request on Terra.

OpenRouter's guidance is to reserve pro mode for hard, high-stakes tasks and keep Terra for routine workloads. Teams with latency budgets should test against the 4.10s P50 figure, and EU-bound deployments should note the missing in-region routing.