GPT-6 Astra

Official

openai/gpt-6-astra

GPT-6 Astra is OpenAI's new flagship, built for the hardest reasoning, coding, and long-running agent work. It takes text and image input, holds a 1.05M-token context window (922K max input), writes up to 128K output tokens, and exposes five reasoning-effort levels from low through max. Prompt caching is supported, and cache reads bill at one tenth of input.

1050K contextVisionReasoningTool callingCodingPrompt cachePaid cache writes

Pricing

Source: openai_2026_09_gpt6_x0.10 · Verified 2026-09-05

InputOfficial $10 / M tokens

$1/ M tokensSave 90%

OutputOfficial $50 / M tokens

$5/ M tokensSave 90%

Cache read $0.1/M · Cache write $1.25/M

Tier	Input	Cache read	Cache write	Output
Standard ≤272K	$1	$0.1	$1.25	$5
Standard >272K	$2	$0.2	$2.5	$7.5

Estimate cost

Input tokensOutput tokens

Estimated cost$0.015

Protocols

OpenAI Chat CompletionsStatus: Available: Streaminghttps://api.onehop.ai/v1
OpenAI ResponsesStatus: Available: Streaminghttps://api.onehop.ai/v1
Anthropic MessagesStatus: Not supported—
Google Vertex AIStatus: Not supported—
OpenAI ImagesStatus: Not supported—
OpenAI SoraStatus: Not supported—

Try it in the chat playground

Call IDs

Use either ID to call this model via the API.

OneHop nameopenai/gpt-6-astra

Usable Model IDs

You can use these names directly in the request model field.

gpt-6openai/gpt-6

Try it

Replace the ONEHOP_KEY placeholder with your API key. Create one →

from openai import OpenAI

client = OpenAI(
    base_url="https://api.onehop.ai/v1",
    api_key="<ONEHOP_KEY>",
)

completion = client.chat.completions.create(
    model="openai/gpt-6-astra",
    messages=[{"role": "user", "content": "What is the meaning of life?"}],
)
print(completion.choices[0].message.content)

base_url: https://api.onehop.ai/v1

Tier

Input

Cache read

Cache write

Output

Standard ≤272K

$0.1

$1.25

Standard >272K

$0.2

$2.5

$7.5

from openai import OpenAI client = OpenAI( base_url="https://api.onehop.ai/v1", api_key="<ONEHOP_KEY>", ) completion = client.chat.completions.create( model="openai/gpt-6-astra", messages=[{"role": "user", "content": "What is the meaning of life?"}], ) print(completion.choices[0].message.content)