← Model list

GPT-5.4

openai·openai/gpt-5.4·BETA·Closed·gpt series

Agent-ready GPT for coding and computer-use workflows at a lower cost

At a glance
Official price$2.50 / $15.00 per 1M
Cache read $0.25 · OpenAI
Lowest paid$0.75 / $6.00
Xpersona Gateway · 6.7× spread
1 more $0 channels
Context1,050,000
⚠ Providers report 272,000–1,050,000; the table below is authoritative
Output limit128,000
Capabilities
ReasoningTool useStructured outputTemperatureAttachments
Modalities
TextImagePDF
Knowledge cutoff2025-08-31
Released / updated2026-03-05 / 2026-03-05

Quality & performanceArtificial Analysis · Intelligence Index v4.1 · rep. tier xhigh

Intelligence53.1
Value26
Coding71.1
Agentic44.2
Output speed131 tok/s
TTFT144.08 s
Cost per task$1.1038
Value formulaIQ 53.1 ÷ min blended $2.063 = 26
Reasoning tier → intelligence / speed (higher tier = stronger but slower)
Non-reasoningIQ 28.3 · 90 tok/s
lowIQ 40.2 · 94 tok/s
xhighIQ 53.1 · 131 tok/s

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 42 providers38 with public prices · 1 free

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
UnoRouter
gpt-5.4:free
GatewayFree1,050,000128,000
XpersonaGateway$0.75$6.00$0.0751,050,000128,000
UnoRouterGateway$1.80$10.801,050,000128,000
PoeGateway$2.20$14.00$0.221,050,000128,000
RequestyGateway$2.48$14.85$0.2471,050,000128,000
AI-ROUTERGateway$2.50$15.00$0.251,050,000128,000
NanoGPTGateway$2.50$15.00$0.25922,000128,000
ImpossiblGateway$2.50
tiered >272K: $5.00
$15.00$0.251,050,000128,000
OpenRouterGateway$2.50
tiered >272K: $5.00
$15.00$0.251,050,000128,000
DaoXEGateway$2.50$15.00$0.251,050,000128,000
OpenAIOfficialFirst-party$2.50
tiered >272K: $5.00
$15.00$0.251,050,000128,000

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Your usage cost

1Xpersona$369.00
2UnoRouter$900.00
3Poe$902.40
4Requesty$970.20
5AI-ROUTER$980.00
6NanoGPT$980.00
12OpenAI · Official$980.00

1 more channels offer $0 (UnoRouter); free tiers usually have rate limits and no SLA, excluded from ranking.

Switch to Xpersona to save $611.00/mo (62%)
Note: this is a gateway; verify availability and rate limits yourself.

Benchmark25 items

NameConditionsScoreMetricSource
SWE-Bench Prodataset: public59.1resolve rateSource ↗
SWE-Atlas Codebase QnAharness: Codex40.8scoreSource ↗
SWE-Atlas Codebase QnAharness: Mini-SWE-Agent36.3scoreSource ↗
SWE-Atlas Refactoringharness: Codex44.29scoreSource ↗
SWE-Atlas Test Writingharness: Codex CLI44.36scoreSource ↗
SWE-Atlas Test Writingharness: Mini-SWE-Agent40scoreSource ↗
Artificial Analysis Coding Agent Indexharness: Codex · variant: medium53.6average pass@1Source ↗
SWE-Atlas Codebase QnAharness: Codex · variant: medium72.4pass@1Source ↗
SWE-Bench Proharness: Codex · variant: medium · dataset: hard-aa18.4pass@1Source ↗
Terminal-Benchharness: Codex · variant: medium · v2.169.8pass@1Source ↗
Artificial Analysis Coding Agent Indexharness: Cursor CLI · variant: medium52.2average pass@1Source ↗
SWE-Atlas Codebase QnAharness: Cursor CLI · variant: medium72.9pass@1Source ↗
SWE-Bench Proharness: Cursor CLI · variant: medium · dataset: hard-aa18.9pass@1Source ↗
Terminal-Benchharness: Cursor CLI · variant: medium · v2.164.7pass@1Source ↗
Terminal-Benchv2.075.1success rateSource ↗
GPQA Diamond92.8accuracySource ↗
Humanity's Last Examvariant: no tools39.8accuracySource ↗
Humanity's Last Examvariant: with tools52.1accuracySource ↗
OSWorld-Verified75success rateSource ↗
BrowseComp82.7accuracySource ↗
GDPval83wins or tiesSource ↗
ARC-AGI-2variant: Verified73.3accuracySource ↗
FrontierMathdataset: Tier 1-347.6accuracySource ↗
FrontierMathdataset: Tier 427.1accuracySource ↗
MMMU Provariant: no tools81.2accuracySource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Reasoning control

effort = loweffort = mediumeffort = higheffort = xhigheffort = maxeffort = nonebudget_tokenseffort = minimal

4 / 42 providers expose no reasoning control (reasoning_options: []).

Experimental modes7 items

ModeProviderInputOutputCache readCache write
fastDatabricks$5.00$30.00$0.50
fastOpenAI$5.00$30.00$0.50
fastMerge Gateway$5.00$30.00$0.50
fastAIHubMix$5.00$30.00$0.50
fastOrcaRouter$5.00$30.00$0.50
fastSnowflake Cortex$5.00$30.00$0.50
fastAnyAPI$5.00$30.00$0.50

Related models

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$15.00$02026-08-052026-08-132026-08-05 · Input list $2.502026-08-06 · Input list $2.502026-08-07 · Input list $2.502026-08-08 · Input list $2.502026-08-09 · Input list $2.502026-08-10 · Input list $2.502026-08-11 · Input list $2.502026-08-12 · Input list $2.502026-08-13 · Input list $2.502026-08-05 · Output list $15.002026-08-06 · Output list $15.002026-08-07 · Output list $15.002026-08-08 · Output list $15.002026-08-09 · Output list $15.002026-08-10 · Output list $15.002026-08-11 · Output list $15.002026-08-12 · Output list $15.002026-08-13 · Output list $15.002026-08-05 · Min blended $2.062026-08-06 · Min blended $2.062026-08-07 · Min blended $2.062026-08-08 · Min blended $2.062026-08-09 · Min blended $2.062026-08-10 · Min blended $2.062026-08-11 · Min blended $2.062026-08-12 · Min blended $2.062026-08-13 · Min blended $2.06