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

GPT-5.6 Sol

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

OpenAI's GPT model.

Top three for

2 tasks, including research, coding and building apps.

Evidence

Independently tested. 29 independent results so far.

Cost

$22 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 Plus$20/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 research2nd of 1178
Best AI for coding and building apps3rd of 1072
Best AI for studying and homework4th of 1279
Best AI for documents, PDFs and notes4th of 1369
Best AI for CVs, job applications and interviews5th of 1453
Best AI for writing6th of 1451
Best AI for maths and science6th of 1276
Best AI for translation and languages8th of 1454
Best AI for planning and personal admin9th of 1442
Best AI for health questions11th of 1433

Test results

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

TestResultPosition
Competition maths
Epoch AI
100%2nd of 12
Research-level maths
Epoch AI
89%3rd of 11
Pulling information out of documents into tables
DTBench authors
96%4th of 13
Would a real maintainer accept its code?
Cognition
47%3rd of 9
People's votes on expert questions
LMArena
rating 15175th of 14
People's votes on following instructions
LMArena
rating 14805th of 14
Building features in real software projects
Datacurve
73%3rd of 8
People's votes on coding questions
LMArena
rating 14936th of 14
People's votes on long, detailed requests
LMArena
rating 14796th of 14
Graduate-level science questions
NYU and others
93%5th of 11
Answering short factual questions correctly
Google DeepMind
70%5th of 11
People's votes on creative writing
LMArena
rating 14507th of 14
People's votes on legal questions
LMArena
rating 14727th of 14
People's votes on writing and language
LMArena
rating 14647th of 14
People's votes on maths questions
LMArena
rating 14727th of 13
People's votes on business and finance questions
LMArena
rating 14568th of 14
People's votes on questions in other languages
LMArena
rating 14508th of 14
People's votes overall
LMArena
rating 14568th of 14
People's votes on back-and-forth conversations
LMArena
rating 14569th of 14
People's votes on health questions
LMArena
rating 143210th of 13
People's votes on science questions
LMArena
rating 145311th of 14

Details

Maker
OpenAI
Released
Date not stated
Status
Current
Replaces
GPT-5.5
Name checked
Confirmed with OpenAI
Intelligence Snacks newsletter

The big AI ideas each week, in your inbox