OpenRouter has listed GPT-6.1 Terra Pro as openai/gpt-6.1-terra-pro, plus a :batch variant. Per OpenRouter, it is the same underlying model as GPT-6.1 Terra, served with reasoning.mode=pro—not a new OpenAI model.

What OpenRouter lists

From the OpenRouter model page and API (/api/v1/models):

  • Ids: openai/gpt-6.1-terra-pro and openai/gpt-6.1-terra-pro:batch
  • What it is: same underlying GPT-6.1 Terra, with reasoning.mode set to pro for higher-quality responses on complex tasks
  • List price: $1 input / $5 output per 1M tokens (same list $/M band as Terra on OpenRouter; the extra cost is from higher pro reasoning-token use, not a higher posted $/M)
  • Context: 1,050,000 tokens (context_length); max completion 128,000
  • Listed: API created 1793038556 → Beijing time 2026-10-27 02:15:56

OpenRouter’s cost note says 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. It is meant for hard, high-stakes problems; everyday coding, agentic, and chat workloads usually stay on gpt-6.1-terra. Batch list pricing is lower and is generally not the everyday default.

How this maps to OpenAI’s API

OpenAI’s Reasoning guide (“Reasoning mode”) says GPT-5.6 and GPT-6 models support standard and pro reasoning modes on the Responses API (standard is the default; pro uses more tokens and latency). Mode is separate from reasoning.effort. OpenRouter’s Terra Pro id is a convenience route that pins pro; it does not mean OpenAI shipped a separate “Terra Pro” model.

Related reading: GPT-6.1 Terra Is Out: Price, Work and Codex Access, and API, GPT-6.1 Terra Is Generally Available in GitHub Copilot, and How Codex Usage Limits Work on Each ChatGPT Plan (October 2026).