Codex
Gpt 6 Astra on AtmoRouter — $0.2250 per million input tokens, $1.13 per million output tokens on a 272K context, through one OpenAI-compatible key.
Input · per Mtok
$0.2250
$10.00 official
Output · per Mtok
$1.13
$50.00 official
Off official
97.8%
vs provider list rate
Context
272K
up to 128K output
Model id
cx/gpt-6-astra
Use this exact id in API calls.
Modality
text,image
Prompt caching supported — cached input $0.0225/Mtok.
Protocols
OpenAI Chat Completions
Anthropic Messages endpoint also available for every model.
curl
curl https://atmorouter.dev/v1/chat/completions \
-H "Authorization: Bearer ar-xxxxxxxxxxxx" \
-H "Content-Type: application/json" \
-d '{
"model": "cx/gpt-6-astra",
"messages": [{"role": "user", "content": "Say hello in one word."}]
}'Python (OpenAI SDK)
from openai import OpenAI
client = OpenAI(
base_url="https://atmorouter.dev/v1",
api_key="ar-xxxxxxxxxxxx",
)
resp = client.chat.completions.create(
model="cx/gpt-6-astra",
messages=[{"role": "user", "content": "Say hello in one word."}],
)
print(resp.choices[0].message.content)Works with anything OpenAI-compatible — Cursor, Cline, LangChain and more. See the cookbook configs at github.com/Micolaabdi/atmorouter-cookbook and the docs for the Anthropic Messages endpoint.
How much does Gpt 6 Astra cost per million tokens?
AtmoRouter bills $0.2250 per million input tokens and $1.13 per million output tokens. The provider's official list rate is $10.00 / $50.00 — AtmoRouter is 97.8% below it.
Does Gpt 6 Astra support streaming?
Yes — server-sent events on the OpenAI-compatible endpoint, and the Anthropic Messages endpoint streams natively.
Is there a subscription or minimum spend?
No. Billing is pay-as-you-go per million tokens, metered from the upstream usage report — never estimated.
Create an account, grab a key, and make your first call in minutes.