gpt-6-astra

Name: gpt-6-astra
Author: Vercel AI Gateway

proprietary

GPT-6 Astra is our most capable model for the most demanding work. Use it for complex reasoning, coding, computer use, research, and document creation. `reasoning.effort` supports `low`, `medium`, `high`, `xhigh`, and `max`.

Intelligence

Reasoning

Speed

Context

1.1M

Max output

128K

Input price

$10/1M tokens

Output price

$50/1M tokens

Capabilities

Show unsupported (2)

Reasoning & Interaction

Reasoning

Tool Calling

Output Control

Structured Output

JSON Mode

Infrastructure

Streaming

Batch Processing

Details

ProviderVercel AI Gateway

CreatorOpenAI

Familygpt-6

Licenseproprietary

Statusactive

Typechat

Input modalitiestext, image

Output modalitiestext

Open weightNo

Parameters—

Architecture—

Context window1.1M tokens

Max input922K tokens

Max output128K tokens

Reasoning tokensYes

Knowledge cutoff2026-04

Release date—

Sourceofficial

Tools

Function Callingfunction_calling

Call external functions and APIs

Pricing

Input

$10

Output

$50

Cache write

—

Cache read

—

Batch in

—

Batch out

—

Family Comparison: gpt-6

Model	Caps	Perf	Speed	Context	Max out	Pricing
gpt-6-luna-decisions	S	—	—	1.1M	—	$0.1$0
gpt-6-luna	RTJJMSB	—	—	1.1M	128K	$0.1$0.5
gpt-6-sol	RTJJMSB	—	—	1.1M	128K	$2$10
gpt-6-astra	RTJJMSB	—	—	1.1M	128K	$10$50

API

GET/v1/models/vercel/openai/gpt-6-astra