GPT-6.1 Terra is an OpenAI model in the GPT-6.1 family, released on 2026-10-26 with a 1.05M-token context window. This page lists what it costs on TokenLab, what OpenRouter lists for the same model, and how to call it. TokenLab publishes this page, so check the numbers against the live pages. Prices are USD per 1 million tokens, read on 2026-11-11: TokenLab from its pricing API, and OpenRouter from the endpoint list of its public API.
Price: TokenLab vs OpenRouter
gpt-6.1-terra | Input | Output | Cache read |
|---|---|---|---|
| TokenLab Verified (prompts up to 272K) | $0.30 | $1.50 | $0.015 |
| TokenLab Verified (prompts over 272K, whole request) | $0.60 | $2.25 | $0.03 |
| Official price (prompts up to 272K) | $1.00 | $5.00 | $0.05 |
| Official price (prompts over 272K, whole request) | $2.00 | $7.50 | $0.10 |
| OpenRouter standard endpoint | $1.00 | $5.00 | $0.05 |
TokenLab Verified is 30% of the Official price on both tiers. Example: across many requests that each stay under 272K input tokens, 10 million input and 2 million output tokens in total cost $6 on TokenLab Verified ($3 + $3) and $20 at the Official price ($10 + $10).
OpenRouter says it passes provider prices through without markup, and charges a fee when you buy credits: 5.5% on its Standard plan (see its pricing page and FAQ). TokenLab charges no fee when you add credit. OpenRouter also lists other endpoints for this model. A flex endpoint and a :batch variant cost $0.50 input and $2.50 output, which is half the standard price. Priority and regional endpoints cost more. TokenLab bills every request at the standard service tier, so compare the standard rows.
Two delivery options exist: Verified is TokenLab's lower-priced supply, and Official is priced at the maker's list price. With Auto, TokenLab tries Verified first and falls back to Official, and you pay for the option that finished the request. See the delivery tiers guide. Web search through the model costs $0.003 per request on Verified and $0.01 on Official.
Limits and what the model supports
From the TokenLab model data: image input, tool use, reasoning and prompt caching. The context window is 1,050,000 input tokens, and the output limit is 128,000 tokens. Accepted request formats are OpenAI Chat Completions and OpenAI Responses. For the current capability flags, read GET /v1/models/gpt-6.1-terra.
Call it in two minutes
from openai import OpenAI
client = OpenAI(
api_key="YOUR_TOKENLAB_KEY",
base_url="https://api.tokenlab.sh/v1",
)
resp = client.chat.completions.create(
model="gpt-6.1-terra",
messages=[{"role": "user", "content": "Summarize this in one sentence: ..."}],
)
print(resp.choices[0].message.content)
Coming from OpenRouter, change the base URL and the key. The model name becomes gpt-6.1-terra instead of openai/gpt-6.1-terra. More on switching in OpenRouter alternatives.
How to test it
Add $5 to $10 of balance, point one non-critical job at the endpoint, and compare cost per finished task and error rate with your current setup. If the numbers do not work for you, stop; nothing else changes. Create a key, see the GPT-6.1 Terra model page or live pricing. Prices can change; the live pages win.
