Codex GPT 6 Astra, 6.1-sol, 6.1-Terra 6-Luna via API

Codex GPT 6 Astra, 6.1-sol, 6.1-Terra 6-Luna via API

A token key with a token package: usage billed as: input + output + cache, multiplied by the model coefficient
Tokens on the token key balance
⚠️ Product format: Token key + Proxy URL
Cache counts 1:1, no long-context surcharge
Personal usage dashboard: statistic.gptclaudegemini.xyz

This is a token key with a token package for key + url connections. It is not a ChatGPT account and not a top-up for your personal OpenAI account.

📦 What You Get
Token key
Tokens on the key balance
Working connection URL
Usage dashboard: remaining package and burn dynamics
Setup instructions and a test-check workflow
Help with first setup

🧠 Models & Burn Coefficients (updated for November 2026)

Each request burns: input + output + cache tokens, all multiplied by k. 1M model tokens burns k × 1M package tokens:
gpt-6-astra - k = 7.5 — 1M astra tokens burns 7.5M credits
gpt-6-sol - k = 2.0 — 1M sol tokens burns 2M credits
gpt-6.1-sol - k = 2.0 — 1M gpt-6.1-sol tokens burns 2M credits
gpt-6.1-terra - k = 1.5 — 1M 6.1-terra tokens burns 1.5M credits
gpt-6-luna - k = 1.0 — 1M luna tokens burns 1M credits
gpt-5.6-terra - k = 2.0 — 1M terra tokens burns 2M credits
gpt-5.5 and codex-auto-review - k = 3.0 — 1M tokens burns 3M credits
gpt-5.4 - k = 3.0
gpt-5.4-mini - k = 1.0
Cache tokens count 1:1 toward the package (they enter the formula like regular tokens). No surcharge for contexts above 272k.

❓FAQ
What is this product?
A token key with a token package via key + url. Choose a package and receive a key and a working Proxy URL; usage is visible in the statistics dashboard.

How is usage billed?
Each request burns (input + output + cache tokens) × k, where k is the model coefficient from the table above. Cache counts 1:1, no long-context surcharge.

How many credits do tokens burn?
1M model tokens burns k × 1M package tokens: astra — 7.5M, 5.6-sol — 3M, 5.5/review — 3M, 5.4 — 3M, 5.6-terra — 2M, 6-sol/6.1-sol — 2M, 6.1-terra — 1.5M, luna/5.4-mini — 1M. Cache counts the same way, 1:1.

How long does access last?
Each package is provided for 30 days.

Where can I see the remaining balance?
In the personal dashboard at statistic.gptclaudegemini.xyz — remaining package and burn dynamics per key.

Can I top up the same key?
Yes, topping up an existing token key is possible — message us in chat after purchase.

Which models are available?
gpt-6-astra, gpt-6-sol, gpt-6.1-sol, gpt-6.1-terra, gpt-6-luna, gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5, gpt-5.4, gpt-5.4-mini, and codex-auto-review.

Is this a ChatGPT account?
No. This is not a ChatGPT account; it is a token key for key + url connections.

Where does it work?
The base setup is Codex CLI. Codex App and compatible clients with key + url access also work.

All coefficients and the period are listed on this page — if anything needs clarifying, message us before buying.
Only the main invocation names are listed here, without a long list of secondary aliases.
If needed, we help finish the setup until it works.

🏷️ Tags
Codex, Token Package, Tokens, Codex CLI, Codex App, API key, proxy api, gpt-6-astra, gpt-6-sol, gpt-6.1-sol, gpt-6.1-terra, gpt-6-luna, gpt-5.5, gpt-5.4, 30 days