IonBin
All pages
Search
Model · OpenAI

GPT models (OpenAI)

OpenAI's GPT models power ChatGPT and the OpenAI API. OpenAI currently points to GPT-6 Astra for complex reasoning and coding, GPT-6.1 Sol to balance capability and cost, and GPT-6 Luna for high-volume work. All three list a 1.05M-token context window and built-in web search, file search and computer-use tools.

Verified Oct 18, 2026 Re-verify: past review date 9 sources

What it is

OpenAI’s GPT family is the model line behind ChatGPT and the OpenAI API. The current flagship tier has three steps (Astra, Sol, Luna). Separate specialised models cover images, voice, transcription and security research.

How it works

On the API you call a model through the Responses API and set a reasoning effort. Higher effort can increase reasoning usage and cost; evaluate its effect on your tasks. Built-in tools let the model search the web, search your uploaded files, call your functions, or operate a computer interface. ChatGPT exposes a separately packaged set of features, models and limits; API specifications do not establish ChatGPT plan entitlements.

Specs and pricing (checked 2026-10-18). All three models list a 1.05M-token context and up to 128K output tokens. USD Standard base text prices per million input/output tokens are Astra $10/$50, Sol $2/$10 and Luna $0.10/$0.50. Above 272K input tokens, the full request uses 2× input/cache rates and 1.5× output rates. Cache writes, tool usage, service tier and regional processing can change the bill. ChatGPT subscriptions are separate; see the ChatGPT record and current plan pricing.

Benefits

  • Different cost tiers with the same advertised API context ceiling. Re-evaluate quality and supported settings before switching.
  • Built-in tools reduce the glue code needed for search and file Q&A.
  • Connects to other tools. OpenAI’s developer docs describe adding a custom MCP server in ChatGPT through its Plugins page, subject to account and workspace policies (see the MCP introduction).
  • Packaged instructions, in transition. Custom GPTs turned a repeatable prompt into a shared tool; OpenAI is moving them to plugins. Its migration guide (checked 2026-10-18) says a migration turns a GPT’s instructions into a skill with its knowledge files and connected apps, that custom actions do not transfer, and that at retirement custom GPTs stop running; it gives no date. Plan the move before you build new ones (see SOPs as custom assistants).

Limits and risks

  • Effort costs money. Match effort to the task and measure the complete workflow, including repeated calls and billable reasoning.
  • Fast-moving lineup. Names, prices and cutoffs here are a dated snapshot.
  • Knowledge cutoff. The cutoff describes training coverage, not guaranteed knowledge. Newer information may also come from supplied files, retrieval or other tools; check the evidence behind each answer (see knowledge cutoff).

Checked specs

Snapshot from official sources on Oct 18, 2026. Model lineups and prices change often; this page is re-verified every 30 days.

Flagship / balanced / high-volumeGPT-6 Astra / GPT-6.1 Sol / GPT-6 Luna [1]
Context window / max output1.05M tokens / 128K tokens (all three) [1]
Knowledge cutoffApr 30, 2026 (Astra, Sol); May 18, 2026 (Luna) [1]
USD Standard base text API prices per 1M tokens (2026-10-18)Astra $10/$50; Sol $2/$10; Luna $0.10/$0.50 input/output. Long-context, cache, tool and service-tier charges differ. [9]
Built-in toolsFunctions, web search, file search, computer use [1]
Reasoning effort settingslow to max (Luna also offers none) [1]
ChatGPT plansSeparate subscriptions and feature limits; see current pricing and ChatGPT record. Go amount not verified in this review. [2]
Custom MCP servers in ChatGPTAdded through ChatGPT Plugins; account and workspace policies apply [3]
Custom GPTs (checked 2026-10-18)Being transitioned to plugins; existing GPTs usable until retirement; custom actions do not transfer; no date on the migration guide [5]

Best-fit stacks

Stack 04

Marketing content & social

Research, draft, design and review content with an assistant and the publishing tools you already use.

Agent compatibility

API accessYesOfficial docs ↗
AppsChatGPT (web, desktop, mobile)
Tool in this entryHow agents connect
ChatGPT Agent platform API/CLIFile import

Levels come from each tool's recorded, sourced compatibility. How the levels work.

Sources

  1. Models · OpenAIofficial accessed Oct 18, 2026
  2. ChatGPT pricing · OpenAIofficial accessed Oct 18, 2026
  3. Connect and test your plugin · OpenAIofficial accessed Oct 18, 2026
  4. What is the Model Context Protocol (MCP)? · MCP projectofficial accessed Oct 18, 2026
  5. Migrate custom GPTs to plugins · OpenAIofficial accessed Oct 18, 2026
  6. gpt-6-astra model details · OpenAIofficial accessed Oct 18, 2026
  7. gpt-6.1-sol model details · OpenAIofficial accessed Oct 18, 2026
  8. gpt-6-luna model details · OpenAIofficial accessed Oct 18, 2026
  9. API pricing · OpenAIofficial accessed Oct 18, 2026

Practitioner views are restated in our words and labeled as views, not measured results. Product facts come from official sources dated above.

Suggest an edit

IonBin's wiki isn't user-edited. Corrections go to our Directory Curator, who checks them against primary sources before anything changes. Spot something wrong or out of date on “GPT models (OpenAI)”? Email us with a link to a primary source.

Suggest an edit by email or write to contact@ionbin.com