GPT-6.1 Terra API: what it is, the Codex catalog and how to call it on QCode
As of 2026-11-05, OpenAI's API changelog says GPT-6.1 Terra (gpt-6.1-terra) was released on 2026-10-26 and can be called through both Responses and Chat Completions; Codex 0.166.0 (2026-10-29) added it to the bundled model catalog. QCode lists gpt-6.1-terra on /models and in its docs; the setup is below.
Updated 2026-11-05
Claude, GPT and Chinese models on one key, billed per token; check /models for callable models.
- Billed per token — live rates on /models
- Pay by card (Visa / Mastercard), Apple Pay, Google Pay or crypto
- Self-service: top up and activate instantly
GPT-6.1 Terra at a glance
Official positioning
OpenAI's model page calls it “A fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.”
Context window (tokens)
Per the official model page: up to 922,000 input tokens and 128,000 output tokens, knowledge cutoff 2026-06-01; it accepts text and images and returns text.
QCode's two OpenAI endpoints
Codex (Responses protocol) uses base_url https://api.qcode.cc/openai, and OpenAI SDKs on Chat Completions use https://api.qcode.cc/openai/v1, both as given in QCode's docs; OpenAI notes that tool calling requires the Responses API.
Codex bundled catalog
The release notes of Codex CLI rust-v0.166.0 (published 2026-10-29) say “Added GPT-6.1 Terra to the bundled, Vertex AI, and Amazon Bedrock model catalogs.”
What GPT-6.1 Terra is, and how it differs from GPT-6.1 Sol
As of 2026-11-05, GPT-6.1 Terra is the API model OpenAI released on 2026-10-26; the changelog describes it as fast, efficient and aimed at visual reasoning, and its model ID is gpt-6.1-terra. The official model page suggests matching product photos to catalog entries, reading technical drawings, acting on application screens and drafting or revising text. GPT-6.1 Sol, from the same generation, is positioned as near-Astra performance at a lower cost for complex coding, computer use and professional work. The two share the same context window, input limit and output limit; they differ in knowledge cutoff (Terra 2026-06-01, Sol 2026-04-30) and in reasoning.effort levels (Terra adds none).
Official and QCode status (checked 2026-11-05)
OpenAI's gpt-6.1-terra model page lists the official prices and notes that cached input is priced at 5% of the uncached input rate; the full price list is on OpenAI's pricing page (developers.openai.com/api/docs/pricing). QCode's /models tags gpt-6.1-terra as NEW with a 922K / 128K window, and the Codex page in its docs calls it a mid-tier GPT-6.1 for everyday coding and batch tasks.
Timeline
The OpenAI API changelog announces GPT-6.1 Terra (gpt-6.1-terra) and points to the GPT-6 guide for prompting and migration guidance.
Codex CLI rust-v0.166.0 is released; GPT-6.1 Terra joins the bundled, Vertex AI and Amazon Bedrock catalogs.
QCode's /models and the Codex page in its docs were checked: both list gpt-6.1-terra.
What the official pages say and what is unpublished
Confirmed (verifiable word for word on official pages)
① Released 2026-10-26 with the model ID gpt-6.1-terra, which is also the default snapshot; ② the Realtime and Fine-tuning endpoints are marked unsupported, as are audio and video modalities; ③ reasoning tokens, streaming, structured outputs, function calling and prompt caching are supported; ④ a request with more than 272K input tokens is billed in full at 2x the input and cache rates and 1.5x the output rate.
Not published officially
The official changelog, the model page and the Codex 0.166.0 release notes give no benchmark scores for Terra and do not say whether it will become Codex's default model; as of 2026-11-05 neither point has been published.
Choosing between Terra and Sol
gpt-6.1-sol
Official positioning: near-Astra performance at a lower cost, for complex coding, computer use and professional work; reasoning.effort runs from low to max; knowledge cutoff 2026-04-30.
gpt-6.1-terra
Official positioning: fast and efficient, for visual reasoning, precise computer use and clear, natural writing; reasoning.effort accepts none, low, medium (default), high, xhigh and max; knowledge cutoff 2026-06-01. Official prices are on OpenAI's pricing page; QCode's rates are on /models.
How to call GPT-6.1 Terra
① Pick an API: OpenAI marks both Chat Completions and Responses as supported for gpt-6.1-terra; use Responses when you need tool calling. ② In Codex: write ~/.codex/config.toml as shown on the Codex page in QCode's docs, with model_provider = "qcode" and model_reasoning_effort = "high" at the top level and name = "qcode", base_url = "https://api.qcode.cc/openai", wire_api = "responses" and requires_openai_auth = true under [model_providers.qcode]. The docs example uses model = "gpt-6.1-sol"; to use Terra, change that line of the example to model = "gpt-6.1-terra" (the docs model table lists it). Put your QCode key, which starts with cr_, in OPENAI_API_KEY inside ~/.codex/auth.json. To switch for a single run, replace the model after codex -m in the docs example with gpt-6.1-terra. ③ For Chat Completions with an OpenAI SDK, set base_url (or the OPENAI_BASE_URL environment variable) to https://api.qcode.cc/openai/v1.
Using GPT-6.1 Terra on QCode
Three steps: register and create a key that starts with cr_; top up with Alipay, WeChat Pay, cards (Visa · Mastercard · Apple Pay · Google Pay) or crypto; set the base_url or OPENAI_BASE_URL above and use gpt-6.1-terra as the model. GPT models use the two OpenAI endpoints and do not work on the Anthropic endpoint. Billing is per token, with per-model rates on /models. Not ready for a monthly plan? Pay-as-you-go Packs start at $4 for a one-time quota you use within 7 days; if you already have a Personal Plan, add a booster pack when you need more quota.
FAQ
What is GPT-6.1 Terra good for?
The official model page suggests matching product photos to catalog entries, reading technical drawings, acting on application screens and drafting or revising text. Complex coding, computer use and professional work are the uses OpenAI lists for GPT-6.1 Sol.
What is the GPT-6.1 Terra API price?
Use the prices listed on OpenAI's pricing page (developers.openai.com/api/docs/pricing) and the gpt-6.1-terra model page; a request over 272K input tokens is billed in full at 2x the input and cache rates and 1.5x the output rate. QCode bills per token, with per-model rates on /models.
Can I pick GPT-6.1 Terra in Codex, and is it the default?
You can pick it: the Codex CLI 0.166.0 release notes (2026-10-29) say it was added to the bundled model catalog. Whether it becomes the default is not stated in the release notes; as of 2026-11-05 that is unpublished.
How does GPT-6.1 Terra relate to GPT-5.6 Terra?
They are two distinct model IDs, gpt-6.1-terra and gpt-5.6-terra. QCode's docs describe the first as a mid-tier GPT-6.1 and the second as the code-optimized GPT-5.6 flagship; the GPT-5.6 tiers are compared at /gpt-5-6-sol-vs-terra-vs-luna.
How do I call GPT-6.1 Terra on QCode?
Use a QCode key: for Codex set base_url to https://api.qcode.cc/openai with wire_api = responses, for Chat use https://api.qcode.cc/openai/v1, and set the model to gpt-6.1-terra; availability follows /models.
How do I top up from China to use GPT-6.1 Terra?
QCode accepts Alipay, WeChat Pay, cards (Visa · Mastercard · Apple Pay · Google Pay) and crypto; Alipay, WeChat Pay and cards usually arrive within a minute, while crypto is checked by hand and credited within one day. A full GPT-6 top-up walkthrough is at /gpt-6-api-china-top-up.
Sources
OpenAI: the API changelog, the gpt-6.1-terra and gpt-6.1-sol model pages (developers.openai.com) and the Codex CLI rust-v0.166.0 release notes (GitHub), all captured on 2026-11-05. QCode: qcode.cc/models, the Codex page and endpoint guide on docs.qcode.cc, and the top-up page, captured the same day.
Call GPT-6.1 Terra with one key
Payment methods and arrival times are on the top-up page; model availability follows /models.
Related reading
Pay-as-you-go Packs
No subscription: buy a quota once and use it within 7 days; extra packs add to the same key.
GPT-6 guide
Release dates, specs and callable models across the GPT-6 generation.
GPT-5.6 Sol vs Terra vs Luna
How the three GPT-5.6 tiers compare and which to default to.
Crypto payment guide
Nine coins, 56 networks, zero platform fee for AI coding API credits.
The details on this page were checked on 2026-11-05 against OpenAI's official pages and QCode's documentation; model availability follows /models, and OpenAI may change these details without notice.