Ordalin
Submit a tool

Reviewed tool profile

GPT-6

OpenAI’s flagship model family: Astra, 6.1 Sol and Luna, with 1.05M-token context.

Sources checked

Visit website ↗
Pricing
API usage is billed per million tokens in USD, with different rates per model and processing tier. Standard short-context rates are $10 input / $50 output for GPT-6 Astra, $2 / $10 for GPT-6.1 Sol and $0.10 / $0.50 for GPT-6 Luna. Long-context requests cost more, Batch and Flex cost half, and Fast and Ultrafast modes cost more. Official source 1 ↗
Available as
Available through the OpenAI API, mainly the Responses API, and OpenAI’s client SDKs. · GPT-6 Astra supports function calling, web search, file search and computer use as built-in tools, with async tool calling while your application runs a tool. Official source 1 ↗ · Official source 2 ↗ Official source 1 ↗ · Official source 2 ↗
Primary group
Coding & Development

What it does

GPT-6 is OpenAI’s current flagship model family for the API, offered in three tiers: GPT-6 Astra for the most demanding reasoning, coding and professional work, GPT-6.1 Sol for near-Astra performance at lower cost, and GPT-6 Luna for fast, high-volume tasks. All three accept text and image input, return text, and have a 1.05M-token context window with up to 128K output tokens.

Official source 1 ↗ · Official source 2 ↗

Features

  • Three tiers for quality, speed and cost. Choose Astra for the hardest work, 6.1 Sol to balance intelligence and cost, or Luna for cheap, high-volume tasks. Official source 1 ↗ · Official source 2 ↗
  • Computer use and multi-step workflows. OpenAI says GPT-6 Astra performs strongly on computer use, browsing, software engineering and science, and carries out multi-step workflows across code, browsers and professional software. Official source 1 ↗
  • 1.05M-token context with image input. Every model in the family accepts text and image input with a 1.05M-token context window and up to 128K output tokens. Official source 1 ↗
  • Async tool calling. GPT-6 can keep reasoning, call other tools or answer independent parts of a request while your application runs a tool. Official source 1 ↗
  • Mid-turn steering. Developers can send extra user instructions, such as corrections or changed requirements, while GPT-6 is still working. Official source 1 ↗
  • Adjustable reasoning effort. Reasoning effort runs from low to max; Luna also supports a no-reasoning setting for the fastest responses. Official source 1 ↗

Pricing & free limits

GPT-6 Astra

$10 input / $1 cached input / $50 outputUSD per million tokens (standard, short context)

Most capable tier. Long context costs $20 input / $75 output; Batch and Flex are half price; Fast and Ultrafast modes cost more. Official source 1 ↗

GPT-6.1 Sol

$2 input / $0.10 cached input / $10 outputUSD per million tokens (standard, short context)

Near-Astra performance at lower cost. Long context costs $4 input / $15 output; Batch is half price. Official source 1 ↗ · Official source 2 ↗

GPT-6 Luna

$0.10 input / $0.01 cached input / $0.50 outputUSD per million tokens (standard, short context)

Fastest and cheapest tier for focused, high-volume tasks. Long context costs $0.20 input / $0.75 output; Batch is half price. Official source 1 ↗ · Official source 2 ↗

Billing details

Batch and Flex processing cost half the standard rate, while Fast mode doubles it and Ultrafast mode (Astra only) costs six times as much. Official source 1 ↗

Regional data-residency and FedRAMP endpoints add a 10% uplift. Official source 1 ↗

Free access & limits

Free access and its limits were not established by the checked official sources.

Best for & limitations

Run complex coding and software-engineering tasks with GPT-6 Astra or GPT-6.1 Sol. Official source 1 ↗ · Official source 2 ↗

Automate multi-step work across browsers and professional software with computer use. Official source 1 ↗

Process large volumes of focused requests cheaply with GPT-6 Luna. Official source 1 ↗

Build agents that search the web, retrieve files and call your own functions across multi-step workflows. Official source 1 ↗ · Official source 2 ↗

  • Tool calling requires the Responses API. For GPT-6.1 Sol, tool calling needs the Responses API; Chat Completions only supports requests without tools, and the none and minimal reasoning settings are not available. Official source 1 ↗
  • Higher per-token price for Astra. Astra costs more per token than the other tiers, though OpenAI says its estimated cost per task was lower than earlier models. Official source 1 ↗ · Official source 2 ↗
  • Knowledge cutoff. Astra and 6.1 Sol have an April 30, 2026 knowledge cutoff and Luna a May 18, 2026 cutoff. Official source 1 ↗
GPT-6 product screenshot
Website previewFixed desktop viewport · 1440 × 900

Similar tools

Published tools that share its categories.