llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeModelsGPT-5.4

OpenAI model product

GPT-5.4

GPT-5.4 is OpenAI's most capable and efficient frontier model for professional work. It combines industry-leading coding capabilities with native computer-use, up to 1M tokens of context, full-resolution vision processing, tool search for large tool ecosystems, and improved reasoning across spreadsheets, presentations, and documents. It is the most token-efficient reasoning model in the GPT-5 series.

Updated Aug 10, 2026. Default version: GPT-5.4

Compare
LLMBoard score75.6GPT-5.4
Coverage100%8 benchmark families
Context window1.1MTokens
Official input price$2.5OpenAI API

On this page

  • Specification
  • Capability
  • Benchmarks
  • Arena
  • Pricing
  • Versions
  • About
  • Compare
  • Similar models
  • FAQ

Model specification

Structured fields from the published default version.

Version
GPT-5.4
Released
Mar 5, 2026
Knowledge cutoff
Aug 31, 2025
Parameters
N/A
Context window
1.1M
Max output
128K
Inputs
image, pdf, text
Outputs
text
Open weights
No
License
Proprietary

Capability profile

This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.

GPT-5.4 category scores

Benchmark results

Published benchmark records for the scored version GPT-5.4.

23 rows
Columns

Show columns

ARC-AGI0.92783.3%CAug 7, 2026
Graphwalks BFS <128k0.922295.2%CAug 7, 2026
Graphwalks parents <128k0.921894.1%CAug 7, 2026
ARC-AGI v20.731686.7%CAug 7, 2026
LiveBench0.833894.6%BAug 7, 2026
Tau2 Telecom1.033594.1%CAug 7, 2026
FrontierMath0.541781.3%CAug 7, 2026
Finance Agent0.66828.6%CAug 7, 2026
Terminal-Bench 2.00.864989.6%CAug 7, 2026
OmniDocBench 1.50.971660.0%CAug 7, 2026
Toolathlon0.583176.7%CAug 7, 2026
FrontierSWE0.591542.9%BAug 7, 2026
Legal Agent Benchmark0.091333.3%BAug 7, 2026
MMMU-Pro0.896587.5%CAug 7, 2026
GPQA0.91123395.7%CAug 7, 2026
OSWorld-Verified0.8112252.4%CAug 7, 2026
DeepSWE 1.10.5131933.3%BAug 7, 2026
Graphwalks parents >128k0.3141823.5%CAug 7, 2026
BrowseComp0.8195868.4%CAug 7, 2026
Graphwalks BFS >128k0.2192214.3%CAug 7, 2026
SWE-Bench Pro0.6204455.8%CAug 7, 2026
MCP Atlas0.7213031.0%CAug 7, 2026
Humanity's Last Exam0.4359262.6%CAug 7, 2026

Arena results

Preference and agent-evaluation signals from published Arena datasets.

100 rows
Columns

Show columns

text factualitykorean51450.5565N/AAug 6, 2026
text factualityfrench81506.91,746N/AAug 6, 2026
text factualityjapanese91463.5422N/AAug 6, 2026
text factualityhard prompts101502.841,194N/AAug 6, 2026
text factualityindustry business and management and financial operations101485.511,083N/AAug 6, 2026
text factualitypolish101460.9763N/AAug 6, 2026
text factualityindustry legal and government111501.53,991N/AAug 6, 2026
text factualityspanish111468.11,419N/AAug 6, 2026
text factualitychinese121526.82,821N/AAug 6, 2026
visiondiagram121319.05,615N/AAug 6, 2026
vision style controldiagram131318.05,615N/AAug 6, 2026
text factualityoverall141475.962,953N/AAug 6, 2026
text factualityexpert141509.44,801N/AAug 6, 2026
text factualitymath141476.32,590N/AAug 6, 2026
text factualitymulti turn141488.69,989N/AAug 6, 2026
documentoverall151470.129,809N/AJul 30, 2026
text factualitycoding151526.116,194N/AAug 6, 2026
text factualitygerman151424.3496N/AAug 6, 2026
text factualityindustry software and it services151511.624,234N/AAug 6, 2026
text factualitynon english151466.833,680N/AAug 6, 2026
document style controloverall161469.329,809N/AJul 30, 2026
text factualityindustry mathematical161475.82,632N/AAug 6, 2026
textkorean181435.11,133N/AAug 6, 2026
text factualityenglish181480.129,164N/AAug 6, 2026
visionhomework181312.22,847N/AAug 6, 2026
text factualityexclude ties191477.148,367N/AAug 6, 2026
vision style controlocr191293.714,870N/AAug 6, 2026
text factualityhard prompts english201503.918,700N/AAug 6, 2026
text factualitylonger query201486.825,428N/AAug 6, 2026
visionoverall201293.020,516N/AAug 6, 2026
visionocr201301.514,870N/AAug 6, 2026
vision style controloverall201280.520,516N/AAug 6, 2026
vision style controlenglish201276.78,737N/AAug 6, 2026
textjapanese211454.0609N/AAug 6, 2026
text factualityindustry writing and literature and language211460.613,568N/AAug 6, 2026
text factualityinstruction following211474.419,951N/AAug 6, 2026
vision style controlhomework211307.02,847N/AAug 6, 2026
text factualityindustry life and physical and social science221490.88,523N/AAug 6, 2026
visionchinese221341.31,059N/AAug 6, 2026
visionenglish221288.78,737N/AAug 6, 2026
text factualityindustry medicine and healthcare241493.43,634N/AAug 6, 2026
text factualityrussian241475.25,427N/AAug 6, 2026
text style controlindustry business and management and financial operations261473.712,871N/AAug 6, 2026
textindustry business and management and financial operations271456.112,871N/AAug 6, 2026
text factualityindustry entertainment and sports and media271441.411,971N/AAug 6, 2026
text style controlindustry writing and literature and language271459.415,139N/AAug 6, 2026
text style controljapanese271450.3609N/AAug 6, 2026
visioncreative writing vision291268.51,149N/AAug 6, 2026
webdevimage to webdev291448.71,802N/AAug 4, 2026
textchinese301509.93,514N/AAug 6, 2026
textnon english301448.333,780N/AAug 6, 2026
vision style controlchinese301299.51,059N/AAug 6, 2026
textindustry writing and literature and language311448.915,139N/AAug 6, 2026
textpolish311454.31,238N/AAug 6, 2026
vision style controlcreative writing vision311253.31,149N/AAug 6, 2026
textexpert321480.95,888N/AAug 6, 2026
text style controlmulti turn321480.211,861N/AAug 6, 2026
text style controlpolish321471.41,238N/AAug 6, 2026
textfrench331470.82,274N/AAug 6, 2026
textrussian331458.46,743N/AAug 6, 2026
textoverall341451.763,097N/AAug 6, 2026
texthard prompts341468.241,311N/AAug 6, 2026
textinstruction following341454.221,300N/AAug 6, 2026
text style controlcoding341513.817,549N/AAug 6, 2026
text style controllonger query341477.626,792N/AAug 6, 2026
text style controlnon english341457.433,780N/AAug 6, 2026
vision style controlhumor341239.6702N/AAug 6, 2026
textindustry legal and government351462.94,965N/AAug 6, 2026
textlonger query351463.026,792N/AAug 6, 2026
textmulti turn351462.211,861N/AAug 6, 2026
text style controlrussian351470.76,743N/AAug 6, 2026
visionhumor351254.5702N/AAug 6, 2026
textexclude ties361450.248,482N/AAug 6, 2026
text style controlfrench361484.82,274N/AAug 6, 2026
text style controlindustry legal and government361475.64,965N/AAug 6, 2026
text style controlspanish361462.51,878N/AAug 6, 2026
text factualitycreative writing371441.18,475N/AAug 6, 2026
text style controlinstruction following371461.221,300N/AAug 6, 2026
text style controlkorean381423.31,133N/AAug 6, 2026
textindustry mathematical391466.53,349N/AAug 6, 2026
textindustry software and it services391474.125,052N/AAug 6, 2026
text style controlexpert391492.05,888N/AAug 6, 2026
text style controlhard prompts391487.541,311N/AAug 6, 2026
textcoding401479.917,549N/AAug 6, 2026
text style controlindustry mathematical401473.03,349N/AAug 6, 2026
texthard prompts english411463.020,026N/AAug 6, 2026
textindustry life and physical and social science411461.510,157N/AAug 6, 2026
textspanish411453.01,878N/AAug 6, 2026
text style controloverall421465.163,097N/AAug 6, 2026
text style controlexclude ties421471.448,482N/AAug 6, 2026
text style controlindustry software and it services431498.825,052N/AAug 6, 2026
textindustry entertainment and sports and media441424.313,399N/AAug 6, 2026
textmath441457.33,334N/AAug 6, 2026
text style controlenglish451468.829,315N/AAug 6, 2026
text style controlhard prompts english451486.820,026N/AAug 6, 2026
text style controlindustry entertainment and sports and media481433.513,399N/AAug 6, 2026
text style controlmath481460.83,334N/AAug 6, 2026
textcreative writing491426.710,176N/AAug 6, 2026
text style controlcreative writing501434.810,176N/AAug 6, 2026
text style controlindustry life and physical and social science501478.410,157N/AAug 6, 2026

Pricing

Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.

Official API
$2.5 input, $15 output per 1M
Official provider
OpenAI
Lowest third-party
From $0.75 input, $6 output per 1M via Xpersona
Tracked offerings
19
18 rows
Columns

Show columns

Xpersonagpt-5.4global$0.75$61.1MAug 7, 2026
UnoRoutergpt-5.4global$1.8$10.81.1MAug 7, 2026
Vivgridgpt-5.4global$2.5$15400KAug 7, 2026
DaoXEgpt-5.4global$2.5$151.1MAug 7, 2026
LLM Gatewaygpt-5.4global$2.5$151.1MAug 7, 2026
OpenCode Zengpt-5.4global$2.5$151.1MAug 7, 2026
GitHub Copilotgpt-5.4global$2.5$151.1MAug 7, 2026
302.AIgpt-5.4global$2.5$151.1MAug 7, 2026
SAP AI Coregpt-5.4global$2.5$151.1MAug 7, 2026
Pioneergpt-5.4global$2.5$151.1MAug 7, 2026
AI-ROUTERgpt-5.4global$2.5$151.1MAug 7, 2026
Abacusgpt-5.4global$2.5$15400KAug 7, 2026
Azure Cognitive Servicesgpt-5.4global$2.5$151.1MAug 7, 2026
Azuregpt-5.4global$2.5$151.1MAug 7, 2026
AIHubMixgpt-5.4global$2.5$151.1MAug 7, 2026
FreeModelgpt-5.4global$2.5$151.1MAug 7, 2026
OpenAIgpt-5.4global$2.5$151.1MAug 7, 2026
Cortecsgpt-5.4global$2.9$15.451.1MAug 7, 2026

Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.

GPT-5.4 versions

All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.

1 rows
Columns

Show columns

GPT-5.4Mar 5, 202675.6N/A1.1M128KNoProprietary

What is GPT-5.4?

A concise description based on the published model registry.

GPT-5.4 is OpenAI's most capable and efficient frontier model for professional work. It combines industry-leading coding capabilities with native computer-use, up to 1M tokens of context, full-resolution vision processing, tool search for large tool ecosystems, and improved reasoning across spreadsheets, presentations, and documents. It is the most token-efficient reasoning model in the GPT-5 series.

Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.

Data snapshot: 2026-08-07. Editorial model content is not available in the backend.

GPT-5.4 vs nearby models

Open a comparison with the three ranked models immediately above and below this model.

GPT-5.4vsGemini 2.5 ProGPT-5.4vsGPT-5.6-LunaGPT-5.4vsGPT-5.5GPT-5.4vsDeepSeek-V3.2GPT-5.4vsQwen3.5 397B A17BGPT-5.4vsNova 2 Pro

Models similar to GPT-5.4

Recommendations prioritize the same model type and family, then the closest published LLMBoard score.

#16+1.1
OP

GPT-5.5

OpenAI

76.8 LLMBoard

DetailsCompare
#15+1.6
OP

GPT-5.6-Luna

OpenAI

77.2 LLMBoard

DetailsCompare
#30-4.0
OP

GPT-5

OpenAI

71.6 LLMBoard

DetailsCompare
#34-6.5
OP

GPT-5.2

OpenAI

69.1 LLMBoard

DetailsCompare
#4+9.8
OP

GPT-5.6-Terra

OpenAI

85.4 LLMBoard

DetailsCompare
#45-12.6
OP

o1

OpenAI

63.0 LLMBoard

DetailsCompare

FAQ

Common questions about GPT-5.4.

When was GPT-5.4 released?

GPT-5.4's default version was released on Mar 5, 2026.

How much does GPT-5.4 cost?

GPT-5.4's official API price is $2.5 per million input tokens and $15 per million output tokens via OpenAI. The lowest tracked third-party offer starts at $0.75 input and $6 output via Xpersona.

Who created GPT-5.4?

GPT-5.4 is published under OpenAI in the model registry.

What is the context window for GPT-5.4?

The default version has a 1.1M token context window.

Is GPT-5.4 open weight?

No. The default version is not marked as having publicly available weights.

How many API providers offer GPT-5.4?

19 published provider offerings are linked to the default version.

What models should I compare GPT-5.4 with?

Nearby ranked alternatives include Gemini 2.5 Pro, GPT-5.6-Luna, GPT-5.5.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai