> GPT-5.4 is a closed AI model from OpenAI, released 5 Mar 2026. Index score 171 (GPT-4 = 100), rank 16 of 229 ranked models. API price $2.5 in / $15 out per million tokens, 1.05M context.

[Models](https://themodelindex.org/models/) /[OpenAI](https://themodelindex.org/labs/openai/)

# GPT-5.4

OpenAI

Released 5 Mar 2026

Closed

[Compare](https://themodelindex.org/compare/gpt-5-4-vs-kimi-k2-5/)

171159–182

Index score

#16 of 229 models

Pro mode: 176 · $30 / $180

8/ 8

Core tests taken

6 of 6 areas · plus 19 anchor tests

Frontier

Versus the frontier at release

set or matched the record

Replaced by [GPT-5.5](https://themodelindex.org/models/gpt-5-5/)

$2.5/ $15

Price per million tokens

input / output, list price

1.05M

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

Among the first models to reach this score.

## Test results · by area · tick = best by any model

Show 13 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

93%

Humanity's Last Exam

36%

Maths

evidence

FrontierMath T1-3

79%

Coding

evidence

SWE-bench Verified

77%

Terminal-Bench

82%

Agents

evidence

APEX-Agents

52%

Novel problems

evidence

ARC-AGI-2

74%

Long tasks

evidence

METR time horizon

5.7 h

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#16

of 219 · 171

LMArena · Maths

maths questions in chat

#23

of 373 · 1491

Epoch Capabilities Index

overall capability from many benchmarks

#16

of 239 · 156.9

LMArena · Text

overall chat quality, judged by people

#30

of 388 · 1475

LMArena · Coding

coding questions in chat

#33

of 383 · 1520

LMArena · Creative writing

creative writing

#49

of 386 · 1442

LMArena · Vision

understanding images

#20

of 149 · 1284

LiveBench · Overall

overall, with fresh questions to limit contamination

#16

of 58 · 78

LMArena · Documents

working with long documents

#13

of 41 · 1471

LMArena · WebDev

building web apps and front ends

#58

of 123 · 1465

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

1st

SWE-bench Verified

Epoch AI · independent

1st

Terminal-Bench

Epoch AI · independent

3rd

working with long documents

LMArena · Documents

3rd

METR time horizon

Epoch AI · independent

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

ChatGPT

app · subscription

↗

OpenAI API

API

↗

#### API providers · 3 · per million tokens, in / out

O

OpenAI

$5.0 / $30

cheapest

A

Azure

$2.75 / $16.5

↗

A

Amazon Bedrock

$2.75 / $16.5

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

5 Mar 2026

Announcement

not linked yet

Release entry

–

Scored as

GPT-5.4

Artificial Analysis index

–

Pro mode

index 176 · $30 / $180 per M

Output speed

–

Hugging Face

closed weights

## OpenAI releases · around this one

ChatGPT

Closed

30 NOV 2022

GPT-4

Closed

14 MAR 2023

index 100

---
Source: https://themodelindex.org/models/gpt-5-4/ · The Model Index · data as of 2026-10-01