Skip to content
ChatGPT API
Documentation

Documentation

Models

Last updated:

Choosing a model

If you're not sure where to start, use gpt-5.6-sol, the new flagship for complex reasoning and coding. For everyday work choose gpt-5.6-terra; for latency and cost pick gpt-5.6-luna or gpt-5.4-mini. All models use the official OpenAI request format and are available via the Responses API and the official client SDKs.

Frontier models

Prices are per 1M tokens, already 50% below official OpenAI rates.

View full pricing
gpt-5.6-sol

GPT-5.6-sol

New

new flagship, best for complex reasoning and coding

Model ID
gpt-5.6-sol
Input price
$2.50 / MTok
Output price
$15.00 / MTok
Latency
Thorough
gpt-5.6-terra

GPT-5.6-terra

New

balanced quality and price for everyday work

Model ID
gpt-5.6-terra
Input price
$1.25 / MTok
Output price
$7.50 / MTok
Latency
Fast
gpt-5.6-luna

GPT-5.6-luna

New

fastest and cheapest of the new family

Model ID
gpt-5.6-luna
Input price
$0.50 / MTok
Output price
$3.00 / MTok
Latency
Fastest
gpt-5.5

GPT-5.5

previous flagship, strong all-round balance

Model ID
gpt-5.5
Input price
$2.50 / MTok
Output price
$15.00 / MTok
Latency
Fast
gpt-5.4

GPT-5.4

solid price/quality baseline

Model ID
gpt-5.4
Input price
$1.25 / MTok
Output price
$7.50 / MTok
Latency
Fast
gpt-5.4-mini

GPT-5.4-mini

fast and cheap workhorse

Model ID
gpt-5.4-mini
Input price
$0.38 / MTok
Output price
$2.25 / MTok
Latency
Faster

Which model should I use?

Pick by the job, not the benchmark. Here's the OpenAI lineup mapped to common tasks - to switch, just pass the model ID in the `model` field.

TaskModelWhy
Hardest reasoning & codinggpt-5.6-solNew flagship, top of the 5.6 family
Everyday work & agentsgpt-5.6-terraBalanced quality and price
Fast, high-volume completionsgpt-5.6-lunaFastest and cheapest of the 5.6 line
Previous-gen flagshipgpt-5.5Strong all-rounder, stable and proven
Mainstream tasks on 5.4gpt-5.4Solid price-to-quality baseline
Cheapest previous-gengpt-5.4-miniFast and inexpensive workhorse

Model FAQ

Which model is best for coding?

For the hardest coding, refactoring and agentic tasks, gpt-5.6-sol is strongest - the new flagship of the 5.6 family. gpt-5.5 remains a solid all-rounder if you want a proven previous-gen model. For fast, cheap day-to-day work, gpt-5.6-luna and gpt-5.4-mini give the best speed-to-cost balance.

What is vibe coding, and which model should I use?

Vibe coding means building software by describing what you want in natural language and letting an AI model write and iterate on the code while you review. Use gpt-5.6-sol for planning and complex changes, and gpt-5.6-terra or gpt-5.4 for day-to-day implementation - connected through Cursor, opencode, Codex CLI or GitHub Copilot by pointing them at our endpoint.

Which OpenAI model is the cheapest?

gpt-5.6-luna is the cheapest and fastest of the new 5.6 line; gpt-5.4-mini is the cheapest previous-gen option. All prices are 50% below the official OpenAI rates - see the pricing page for the full per-model table.

How do I switch between models?

Change the model ID in the standard `model` field of your request. Everything else - your key, base URL, SDK and code - stays the same, so switching models is a one-line change.