GPT-6 Luna

Name: GPT-6 Luna model facts
Creator: OSMEXO

openai/gpt-6-luna

OpenAI
Family · GPT-6 Luna
Released 22 Sep 2026
File · Image · Text → Text
Reasoning · adjustable
Tools · JSON · Vision

Live on OSMEXO

Try now Request a discount

API setup & price →

Spec & list price

Context1M

Max output128K

Input / 1M$0.10

Output / 1M$0.505× input

Scores · standing

Intelligence38.1#45/489

LiveBench64.6#17/32

Usage

Tokens / day814.7B▲ 13% 7d

Usage rank#5

Requests / day42.7M

Tool calls / req0.24

Per request · 30 days

Avg input17.4K1.2× median

Avg output5320.8× median

Cache hit79%1.1× median

Reasoning share56%about median

Tool-call failures0.5%0.3× median

Consumption

Peak 2.8T 29 OctLow 591.2B 27 SepRank today #5

News & trends

52 mentions this week ▲ 73%this model · videos, posts, articles · 30 days

Trending · 14 days

News · latest from OpenAI

GPT-6.1 Terra vs. Claude Opus 5.5: Cheaper and faster in three tests at maxthenewstack.io · 2 Nov
Chinese labs now target OpenAI's GPT-6 Astra for distillationfork.zone · 1 Nov
GPT-6.1 Terrasimonwillison.net · 31 Oct
GPT-5.6 Sol can beat Factorio on peaceful moderyanmadden.net · 31 Oct

Benchmarks

bar = share of the top score · Benchmark data: Artificial Analysis

Long-context reasoning#15 of 393

83.3%

SciCode#35 of 160

54.6%

Intelligence index#45 of 489

38.1

Humanity's Last Exam#59 of 465

38.5%

Terminal-Bench 4.0#41 of 152

12.6%

Speed · median across 7 providers, latest day

Output speed#140/33476 tok/s1.2× faster than median

Fastest provider120 tok/sbest of 7 providers

First token#2551.89 s1.8× longer than median

First answer token#9591 sthinking included

#4 cheapest of 16 with a similar score

Value

Intelligence Index

Blended price per 1M tokens (log)

LiveBench suites

Browser70.3

Coding84.5

Data Analysis73.4

Design82.5

Long Agentic30.9

Reasoning81.8

Research43.5

▲ Strongest: Coding, Design. ▼ Weakest: Long Agentic (31).

Technical details

Reasoning: on · medium
Tokenizer: GPT
Moderation: moderated

Reasoning effort

max
xhigh
high
medium · default
low
none

Request parameters 11

tools
tool_choice
response_format
structured_outputs
reasoning
reasoning_effort
include_reasoning
max_tokens
max_completion_tokens
seed

+1 more

verbosity

Interesting facts

Less output room than similar models (128.0K against a 734.0K median).
Released 41 days ago.

When to use

Best for

Cost-aware coding and design drafting
Very long document question answering
Reasoning with adjustable effort
Image and file understanding in apps
Fast tool-calling assistants

Think twice for

Research-heavy deep dives
Terminal-heavy autonomous agents
Self-hosting (closed weights)

Research papers

Same family