Skip to main content
ModelScale

GPT-5.4 Pro

OpenAI · Proprietary · rank 60 · bench-align-v5

Canonical idgpt-5-4-pro
Overall score60.27
Context window1.05M tokens
Release date2026-03-05
Access typeProprietary
Blended $/1M$67.5075% input / 25% output

Capability shape

Seven axes from the ranking source. A missing axis is drawn as a gap.

Capability evidence

Agentic79.3
CodingUnavailable
Knowledge81.7
Reasoning70.1
Multimodal & GroundedUnavailable
Instruction FollowingUnavailable
Math69

Runtime service evidence

Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim.

Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.

Evidence key: Observed

Runtime measurements
MeasurementValueObservedLast goodEvidence
Time to first token151.79 s2026-09-172026-09-17
Throughput74 tok/s2026-09-172026-09-17

Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.

Endpoint and price matrix

Every published price component, including cache reads and writes.

Price components
ComponentUSDEvidence
Input / 1M tokens$30.00
Output / 1M tokens$180.00
Cache read / 1M tokensUnavailable
Cache write / 1M tokensUnavailable
Blended / 1M (75% input / 25% output)$67.50
Self-hosted listingOpenAI's current API pricing page lists GPT-5.4 Pro at $30.00 input / $180.00 output per million tokens for short-context requests, with higher pricing for long-context requests and no cached-input rate.The rates above are a hosted price matched from another provider, not a first-party list price.

Workload-aware monthly cost example

10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. This uses the same calculator as the cost simulator, so an unavailable applicable rate makes the example unavailable too.

Modelled monthly cost$190.08
Modelled tokens2.82M
Open the simulatorChange this workload

Benchmark record

11 matched benchmark rows with their published value, unit, and provenance.

Knowledge

Knowledge benchmarks
BenchmarkValueTasksDifficultyProvenance
HLE
Humanity's Last Exam
58.7Expert-level questionsFrontier expert levelHumanity's Last Exam
FrontierScience
FrontierScience
36.7Research-level science tasksResearch frontierFrontierScience
FrontierScience Research
FrontierScience Research
36.7Scientific research problemsFrontier scientific researchMuse Spark Eval Methodology
HLE w/o tools
Humanity's Last Exam without tools
42.7Expert-level questionsFrontier expert levelIntroducing GPT-5.4 mini and nano

Mathematics

Mathematics benchmarks
BenchmarkValueTasksDifficultyProvenance
IPhO 2025 (Theory)
International Physics Olympiad 2025 (Theory)
93.53 olympiad theory problemsInternational olympiad physicsMuse Spark Eval Methodology
FrontierMath (legacy)
FrontierMath legacy aggregate
50Historical aggregateResearch-level mathematicsFrontierMath: A Benchmark for Evaluating Advanced Mathematical Reasoning in AI
FrontierMath v2 (Tiers 1-3)
FrontierMath v2 Tiers 1-3
50.000295 private advanced mathematics problemsFrom olympiad-plus to early research mathematicsFrontierMath v2 benchmark hub
FrontierMath v2 (Tier 4)
FrontierMath v2 Tier 4
37.50043 private extreme-difficulty mathematics problemsResearch-level mathematics requiring hours or days of expert workFrontierMath Tier 4 v2 leaderboard

Reasoning

Reasoning benchmarks
BenchmarkValueTasksDifficultyProvenance
ARC-AGI-2
Abstraction and Reasoning Corpus for AGI v2
83.3Visual pattern completion and abstract reasoningExpert-level — hardest public reasoning benchmarkARC-AGI-2: A Harder General Intelligence Benchmark
CritPt
Critical Physics Tasks
30.0Research-level physics questionsResearch-level physics reasoningCritPt Benchmark Leaderboard

Agentic

Agentic benchmarks
BenchmarkValueTasksDifficultyProvenance
BrowseComp
BrowseComp
89.3Research questions requiring browsingHard web researchBrowseComp

Lifecycle and limitations log

Lifecycle events the source associates with this model, plus what this profile does not claim.

No lifecycle event references this model.

That is not evidence the model has no lifecycle plan — only that this source published none.

What this profile does not claimValues are reproduced exactly as their sources published them, in the units those sources declared; none are converted, interpolated, or averaged across providers. Any field marked unavailable was attempted and not returned — the attempt timestamp is in each badge. Last attempted fetch for this model’s score: 2026-09-17 17:17 UTC.