PodcastIntelligence Snacks · Weekly conversations about AI, software and business, from the team behind Normie Mode.Listen →
Models · OpenAI

GPT-5.6 Luna

A newer version exists (GPT-6 Luna), but this is still what some apps give you, so we keep comparing it.

OpenAI's GPT model.

Top three for

No task yet.

Evidence

Independently tested. 23 independent results so far.

Cost

$1.24 per 1,000 typical requests, if you use it directly.

Reads at once

1,575 pages of about 500 words.

Where you can use it

The apps and plans that include it, cheapest first.

App and planPriceAccess
ChatGPT FreeFreeIncluded, the default
ChatGPT Go$8/monthIncluded, the default

Where it ranks, task by task

Its place on each task's score, among the models we compare that have enough results to score.

TaskPositionTask score
Best AI for coding and building apps7th of 1049
Best AI for maths and science8th of 1267
Best AI for studying and homework9th of 1267
Best AI for research9th of 1140
Best AI for documents, PDFs and notes11th of 1336
Best AI for CVs, job applications and interviews12th of 1422
Best AI for planning and personal admin13th of 1421
Best AI for translation and languages13th of 1430
Best AI for writing13th of 1420
Best AI for health questions13th of 1411

Test results

Every test we use that has a result for this exact version.

TestResultPosition
Research-level maths
Epoch AI
82%4th of 11
Competition maths
Epoch AI
98%6th of 12
Building features in real software projects
Datacurve
67%5th of 8
Graduate-level science questions
NYU and others
92%8th of 11
People's votes on maths questions
LMArena
rating 145510th of 13
Would a real maintainer accept its code?
Cognition
40%7th of 9
People's votes on business and finance questions
LMArena
rating 144011th of 14
People's votes on expert questions
LMArena
rating 147911th of 14
People's votes on legal questions
LMArena
rating 144711th of 14
Answering short factual questions correctly
Google DeepMind
41%9th of 11
Pulling information out of documents into tables
DTBench authors
89%11th of 13
People's votes on coding questions
LMArena
rating 146612th of 14
People's votes on following instructions
LMArena
rating 144012th of 14
People's votes on long, detailed requests
LMArena
rating 143712th of 14
People's votes on health questions
LMArena
rating 139612th of 13
People's votes on back-and-forth conversations
LMArena
rating 143213th of 14
People's votes on creative writing
LMArena
rating 139713th of 14
People's votes on questions in other languages
LMArena
rating 141813th of 14
People's votes overall
LMArena
rating 143213th of 14
People's votes on science questions
LMArena
rating 143213th of 14
People's votes on writing and language
LMArena
rating 142013th of 14

Details

Maker
OpenAI
Released
Date not stated
Status
Current
Name checked
Not yet confirmed with the maker
Intelligence Snacks newsletter

The big AI ideas each week, in your inbox