Models/GPT-6 Astra (Max)

GPT-6 Astra (Max)Ready

gpt-6-astra

Serverless APILLMReasoningVisionPrompt cachingClosed

A frontier multimodal model oriented toward deep research and complex tool orchestration. It ranks #5 of the top 20 frontier models on the Artificial Analysis Intelligence Index with a score of 52.7. Released September 3, 2026.

Context

—

Parameters

Intelligence rank · 52.7

Closed

Weights

View full spec

ReleasedSep 3, 2026

Knowledge cutoffApr 30, 2026

Size classlarge

Input modalitiestext, image

Output modalitiestext

Median output speed57 tok/s

Median time to first token269.3s

Usage guide

1. Choose how you'd like to use GPT-6 Astra (Max)

☁️ Serverless

Use with pay-per-token pricing. Zero setup, no cold starts, automatic failover between upstream routes.

Usage tiers · Standard • Priority

Try it — sign in

curl https://tokenhub-staging.coality.io/v1/chat/completions \
  -H "Authorization: Bearer $TOKEN_HUB_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-6-astra",
    "messages": [{"role": "user", "content": "Explain vector databases in one paragraph"}]
  }'

Pricing

Bucket	Price	Notes
Input tokens	$12.50	per 1M tokens
Output tokens	$62.50	per 1M tokens
Cache read	$1.25	cached input · 90% off
Cache write	$15.63	per 1M tokens

Billed per token with micro-dollar precision · effective Oct 26, 2026 · default key limits 600 req/min · 200K tokens/min

Benchmarks

GPQA Diamond

96.1

Humanity's Last Exam

54.7

MMMU-Pro

86.9

SciCode

56.5

Terminal-Bench 4.0

59.1

Long Context Reasoning

80.7

Frontier Code

47.0

Scores from the Artificial Analysis evaluation suite (0–100 scale).

About this model

Closed model, with a 1M context window, native image input, visible reasoning traces, prompt caching at a 90% discount on cached input, 57 output tokens/s median serving speed and 269.3s to first token.