Model updates

GPT-6.1 Terra is now available on BetterToken

Use gpt-6.1-terra through the BetterToken Responses API. Review official specifications, tool calling, setup, and current cache billing.

Updated: October 27, 2026

GPT-6.1 Terra is now available on BetterToken as gpt-6.1-terra. OpenAI positions it for efficient visual reasoning, computer use, and natural writing. Start with a writing or tool-calling task to evaluate its results.

Official specifications

Input and output: text and image input; text output.
Context: a 1,050,000-token context window and up to 128,000 output tokens.
Reasoning effort: reasoning.effort accepts none, low, medium (default), high, xhigh, or max.

These are OpenAI specifications. Image input and very long contexts have not completed dedicated validation on BetterToken. Your client and access path may impose lower limits.

Connect through Responses

BetterToken has verified streaming and non-streaming Responses requests, function calls, and continued conversations after submitting tool results. Use the Responses API for tool calling.

Use your existing BetterToken API Key, or register and create one.
Enter that key in your client’s API Key field. Set the Base URL to https://bettertoken.ai/v1 and Model ID to gpt-6.1-terra.
Select the Responses protocol, using POST /v1/responses. For direct requests, set the Authorization: Bearer <BETTERTOKEN_API_KEY> header.
Test a representative task before switching an existing workflow.

See the Responses API for request parameters and the Codex CLI setup guide for client configuration.

Pricing and caching

Check the model plaza for current USD and RUB prices. BetterToken currently bills ordinary input and cache writes at the ordinary input rate. OpenAI’s cache-write premium is not charged to you; BetterToken covers the difference. Cache reads have a separate rate.

Requests exceeding 272,000 input tokens use the long-context tier for the entire request. Check the model plaza for rates.