# GPT-5.4 Pro Benchmark Scores & Performance

> GPT-5.4 Pro by OpenAI scores 61.75/100 overall, ranking #60 out of 411 AI models.

Canonical page: https://benchlm.ai/models/gpt-5-4-pro

Last updated: September 4, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | OpenAI |
| Source Type | Proprietary |
| Reasoning Type | Reasoning |
| Context Window | 1.05M |
| Overall Score | 61.75/100 |
| Overall Rank | #60 of 411 |

## Family & Coverage

- Family: GPT-5.4
- Variant: pro
- Benchmarks covered: 11 of 422
- Sibling models: [GPT-5.4](/models/gpt-5-4), [GPT-5.4 mini](/models/gpt-5-4-mini), [GPT-5.4 nano](/models/gpt-5-4-nano)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score |
|-----------|-------|
| [BrowseComp](/benchmarks/browsecomp) | 89.3% |

## Reasoning Benchmarks

| Benchmark | Score |
|-----------|-------|
| [ARC-AGI-2](/benchmarks/arc-agi-2) | 83.3% |
| [CritPt](/benchmarks/critpt) | 30.0% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [HLE](/benchmarks/hle) | 58.7% |
| [FrontierScience](/benchmarks/frontierscience) | 36.7% |
| [FrontierScience Research](/benchmarks/frontierscienceresearch) | 36.7% |
| [HLE w/o tools](/benchmarks/hlenotools) | 42.7% |

## Mathematics Benchmarks

| Benchmark | Score |
|-----------|-------|
| [IPhO 2025 (Theory)](/benchmarks/ipho2025theory) | 93.5% |
| [FrontierMath (legacy)](/benchmarks/frontiermath) | 50% |
| [FrontierMath v2 (Tiers 1-3)](/benchmarks/frontiermathv2tiers13) | 50.000% |
| [FrontierMath v2 (Tier 4)](/benchmarks/frontiermathv2tier4) | 37.500% |

## Other OpenAI Models

- [GPT-6 Astra](/models/gpt-6-astra) - Score: 81.05
- [GPT-5.6 Sol](/models/gpt-5-6-sol) - Score: 79.64
- [GPT-5.5](/models/gpt-5-5) - Score: 73.27
- [GPT-5.6 Terra](/models/gpt-5-6-terra) - Score: 71.44
- [GPT-5.4](/models/gpt-5-4) - Score: 70.96
- [GPT-5.2 Pro](/models/gpt-5-2-pro) - Score: 68.73
- [GPT-5.6 Luna](/models/gpt-5-6-luna) - Score: 65.54
- [GPT-5.2](/models/gpt-5-2) - Score: 65.45
- [GPT-5.3 Codex](/models/gpt-5-3-codex) - Score: 65.22
- [GPT-5.1](/models/gpt-5-1) - Score: 63.68
- [GPT-5.5 Pro](/models/gpt-5-5-pro) - Score: 63.07
- [GPT-5.4 mini](/models/gpt-5-4-mini) - Score: 62.63
- [GPT-5.4 nano](/models/gpt-5-4-nano) - Score: 62.19
- [GPT-5.2 Instant](/models/gpt-5-2-instant) - Score: 58.4
- [GPT-5.3 Instant](/models/gpt-5-3-instant) - Score: 58.3
- [GPT-5 (high)](/models/gpt-5-high) - Score: 58.03
- [GPT-5.2-Codex](/models/gpt-5-2-codex) - Score: 56.38
- [GPT-5.3-Codex-Spark](/models/gpt-5-3-codex-spark) - Score: 56.34
- [GPT-5.1-Codex-Max](/models/gpt-5-1-codex-max) - Score: 53.95
- [GPT-5 (medium)](/models/gpt-5-medium) - Score: 52.81
- [GPT-5.1-Codex](/models/gpt-5-1-codex) - Score: 51.62
- [o4-mini (high)](/models/o4-mini-high) - Score: 49.57
- [GPT-OSS 120B](/models/gpt-oss-120b) - Score: 49.26
- [o3-pro](/models/o3-pro) - Score: 49.16
- [o3](/models/o3) - Score: 49.12
- [o3-mini](/models/o3-mini) - Score: 48.06
- [o1](/models/o1) - Score: 47.89
- [o1-preview](/models/o1-preview) - Score: 47.79
- [GPT-5 mini](/models/gpt-5-mini) - Score: 46.29
- [o1-pro](/models/o1-pro) - Score: 45.5
- [GPT-4.1](/models/gpt-4-1) - Score: 43.81
- [GPT-OSS 20B](/models/gpt-oss-20b) - Score: 42.95
- [GPT-4o](/models/gpt-4o) - Score: 40.48
- [GPT-5 nano](/models/gpt-5-nano) - Score: 39.72
- [GPT-4.1 mini](/models/gpt-4-1-mini) - Score: 34.46
- [GPT-4o mini](/models/gpt-4o-mini) - Score: 34.21
- [GPT-4.1 nano](/models/gpt-4-1-nano) - Score: 28.43
- [GPT-4 Turbo](/models/gpt-4-turbo) - Score: 25.47
- [GPT-5.6 Cyber](/models/gpt-5-6-cyber) - Score: not computed
- [GPT Realtime 1.5](/models/gpt-realtime-1-5) - Score: not computed
- [GPT-4o mini TTS](/models/gpt-4o-mini-tts) - Score: not computed
- [GPT Realtime](/models/gpt-realtime) - Score: not computed
- [GPT Realtime 2](/models/gpt-realtime-2) - Score: not computed
- [GPT Realtime mini](/models/gpt-realtime-mini) - Score: not computed
- [GPT-4o Audio](/models/gpt-4o-audio) - Score: not computed
- [GPT-4o mini Audio](/models/gpt-4o-mini-audio) - Score: not computed