# GPT-5.4

OpenAI · Proprietary · rank 12 · bench-align-v5

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelscale.dev/models/gpt-5-4  
JSON: https://modelscale.dev/api/model/gpt-5-4

## Facts

| Field | Value | Evidence |
| --- | --- | --- |
| Canonical id | `gpt-5-4` | — |
| Overall score | 71.16 | Observed 2026-09-22 · source benchlm:models |
| Context window | 1.05M tokens | Observed 2026-09-22 · source benchlm:models |
| Release date | 2026-03-05 | Observed 2026-09-22 · source benchlm:models |
| Access type | Proprietary | — |
| Blended $/1M (75% input / 25% output) | $5.63 | Derived from the input and output rates below |

## Capability evidence

Seven axes from the ranking source. An axis the source did not score is unavailable, not zero.

| Axis | Score | Evidence |
| --- | --- | --- |
| Agentic | 72 | Observed 2026-09-22 · source benchlm:models |
| Coding | 42.7 | Observed 2026-09-22 · source benchlm:models |
| Knowledge | 76.2 | Observed 2026-09-22 · source benchlm:models |
| Reasoning | 57.2 | Observed 2026-09-22 · source benchlm:models |
| Multimodal & Grounded | 69.3 | Observed 2026-09-22 · source benchlm:models |
| Instruction Following | 89.2 | Observed 2026-09-22 · source benchlm:models |
| Math | 64.4 | Observed 2026-09-22 · source benchlm:models |

## Runtime service evidence

Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim. Regional or per-endpoint measurements appear only when the API supplies them; none are modelled.

| Measurement | Value | Observed | Last good | Evidence |
| --- | --- | --- | --- | --- |
| Time to first token | 150.63 s | 2026-09-22 | 2026-09-22 | Observed 2026-09-22 · source benchlm:speed |
| Throughput | 143 tok/s | 2026-09-22 | 2026-09-22 | Observed 2026-09-22 · source benchlm:speed |

## Endpoint and price matrix

Every published price component, including cache reads and writes.

| Component | USD | Evidence |
| --- | --- | --- |
| Input / 1M tokens | $2.50 | Observed 2026-09-22 · source benchlm:pricing |
| Output / 1M tokens | $15.00 | Observed 2026-09-22 · source benchlm:pricing |
| Cache read / 1M tokens | $0.25 | Observed 2026-09-22 · source benchlm:pricing |
| Cache write / 1M tokens | Unavailable | Unavailable · source openrouter:pricing |
| Blended / 1M (75% input / 25% output) | $5.63 | Derived — from the input and output rates above; it has no source record of its own |
| Cost per successful task (LiveBench) | Unavailable | Unavailable · source livebench:table |

**Self-hosted listing.** OpenAI's current API pricing page lists GPT-5.4 at $2.50 input / $0.25 cached input / $15.00 output per million tokens for short-context requests, with higher pricing for long-context requests. 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. Derived here from the published rates above by this site's own calculator — not a figure any source published.

| Field | Value |
| --- | --- |
| Modelled monthly cost | $15.84 |
| Modelled tokens | 2.82M |

## Benchmark record

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

### Knowledge

| Benchmark | Value | Tasks | Difficulty | Provenance |
| --- | --- | --- | --- | --- |
| GPQA (Graduate-Level Google-Proof Q&A) | 92.8 | 448 questions | Graduate level | [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://arxiv.org/abs/2311.12022) |
| GPQA-D (GPQA Diamond) | 92.8 | Graduate-level science questions | Graduate level | [Trinity-Large-Thinking: Scaling an Open Source Frontier Agent](https://www.arcee.ai/blog/trinity-large-thinking) |
| HLE (Humanity's Last Exam) | 52.1 | Expert-level questions | Frontier expert level | [Humanity's Last Exam](https://lastexam.ai/) |
| Artificial Analysis Intelligence Index | 39.0 | Cross-benchmark intelligence index | Display-only external reference | [Artificial Analysis](https://artificialanalysis.ai/) |
| AA-GPQA Diamond (Artificial Analysis GPQA Diamond) | 92.0 | Graduate-level science questions | Graduate-level science reasoning | [Artificial Analysis GPQA Diamond Benchmark Leaderboard](https://artificialanalysis.ai/evaluations/gpqa-diamond) |
| AA-HLE (Artificial Analysis Humanity's Last Exam) | 43.7 | Expert-level questions | Frontier expert reasoning | [Artificial Analysis Humanity's Last Exam Benchmark Leaderboard](https://artificialanalysis.ai/evaluations/hle) |
| AA-Omniscience Index (Artificial Analysis Omniscience Index) | 5.8 | Knowledge questions | Broad factual knowledge | [AA-Omniscience: Knowledge and Hallucination Benchmark](https://artificialanalysis.ai/evaluations/omniscience) |
| AA-Omniscience Accuracy (Artificial Analysis Omniscience Accuracy) | 50.8 | Knowledge questions | Broad knowledge | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/grok-4-3) |
| AA-Omniscience Hallucination Rate (Artificial Analysis Omniscience Hallucination Rate) | 91.7 | Knowledge questions | Factuality | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/grok-4-3) |
| HealthBench Hard | 40.1 | 1,000 health prompts | Advanced health reasoning | [Muse Spark Eval Methodology](https://ai.meta.com/static-resource/muse-spark-eval-methodology) |
| HealthBench Professional | 48.1 | Clinician chat tasks | Professional clinical workflows | [HealthBench Professional: Evaluating Large Language Models on Real Clinician Chats](https://arxiv.org/abs/2604.27470) |
| MedXpertQA (Text) (MedXpertQA Text) | 59.6 | 2,450 medical multiple-choice questions | Professional medical knowledge | [Muse Spark Eval Methodology](https://ai.meta.com/static-resource/muse-spark-eval-methodology) |
| HLE w/o tools (Humanity's Last Exam without tools) | 39.8 | Expert-level questions | Frontier expert level | [Introducing GPT-5.4 mini and nano](https://openai.com/index/introducing-gpt-5-4-mini-and-nano/) |

### Coding

| Benchmark | Value | Tasks | Difficulty | Provenance |
| --- | --- | --- | --- | --- |
| LiveCodeBench Pro | 87.5 | Quarter-specific contest programming sets | High-end contest programming | [LiveCodeBench Pro: How Do Olympiad Medalists Judge LLMs in Competitive Programming?](https://arxiv.org/abs/2506.11928) |
| SWE-bench Pro | 57.7 | 1,865 repository problems | Long-horizon professional engineering | [SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?](https://arxiv.org/abs/2509.16941) |
| Vibe Code Bench (Vibe Code Bench v1.1) | 67.42 | End-to-end web application builds | End-to-end software delivery | [Vibe Code Bench: Evaluating AI Models on End-to-End Web Application Development](https://www.vals.ai/benchmarks/vibe-code) |
| React Native Evals | 85.3 | React Native app implementation tasks | Production mobile app engineering | [React Native Evals](https://rn-evals.vercel.app/) |
| AA Coding Index (Artificial Analysis Coding Index) | 71.0 | Cross-benchmark coding index | Display-only external reference | [Artificial Analysis model leaderboards](https://artificialanalysis.ai/leaderboards/models) |

### Mathematics

| Benchmark | Value | Tasks | Difficulty | Provenance |
| --- | --- | --- | --- | --- |
| FrontierMath v2 (Tiers 1-3) (FrontierMath v2 Tiers 1-3) | 47.600 | 295 private advanced mathematics problems | From olympiad-plus to early research mathematics | [FrontierMath v2 benchmark hub](https://epoch.ai/benchmarks/frontiermath-tier-4-v2) |
| FrontierMath v2 (Tier 4) (FrontierMath v2 Tier 4) | 27.100 | 43 private extreme-difficulty mathematics problems | Research-level mathematics requiring hours or days of expert work | [FrontierMath Tier 4 v2 leaderboard](https://epoch.ai/benchmarks/frontiermath-tier-4-v2?view=graph&tab=leaderboard) |

### Reasoning

| Benchmark | Value | Tasks | Difficulty | Provenance |
| --- | --- | --- | --- | --- |
| ARC-AGI-2 (Abstraction and Reasoning Corpus for AGI v2) | 73.95 | Visual pattern completion and abstract reasoning | Expert-level — hardest public reasoning benchmark | [ARC-AGI-2: A Harder General Intelligence Benchmark](https://arcprize.org/arc-agi/2/) |
| ARC-AGI-3 (Abstraction and Reasoning Corpus for AGI v3) | 0.21 | Interactive game-like tasks with hidden rules | Frontier agentic reasoning | [ARC-AGI-3: A New Challenge for Frontier Agentic Intelligence](https://arcprize.org/media/ARC_AGI_3_Technical_Report.pdf) |
| AA-LCR (Artificial Analysis Long Context Reasoning) | 82.0 | Long-context reasoning tasks | Long-context reasoning | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/grok-4-3) |
| CritPt (Critical Physics Tasks) | 23.4 | Research-level physics questions | Research-level physics reasoning | [CritPt Benchmark Leaderboard](https://artificialanalysis.ai/evaluations/critpt) |

### Instruction Following

| Benchmark | Value | Tasks | Difficulty | Provenance |
| --- | --- | --- | --- | --- |
| AA-IFBench (Artificial Analysis IFBench) | 73.9 | Verifiable instruction constraints | Instruction precision | [Artificial Analysis IFBench Benchmark Leaderboard](https://artificialanalysis.ai/evaluations/ifbench) |

### Agentic

| Benchmark | Value | Tasks | Difficulty | Provenance |
| --- | --- | --- | --- | --- |
| Terminal-Bench 2.0 | 75.1 | Terminal-based software tasks | Professional software engineering | [Terminal-Bench 2.0](https://www.tbench.ai/) |
| BrowseComp | 82.7 | Research questions requiring browsing | Hard web research | [BrowseComp](https://openai.com/index/browsecomp/) |
| GDPval-AA | 1307 | Agentic real-world work tasks | Professional agentic workflows | [DeepSeek-V4 Technical Report](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/main/DeepSeek_V4.pdf) |
| GDPval-AA (GDPval-AA normalized) | 36.6 | Economically valuable tasks | Professional agentic workflows | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/grok-4-3) |
| APEX-Agents-AA | 33.3 | 452 professional-services agent tasks | Long-horizon workplace agent tasks | [APEX-Agents-AA Benchmark Leaderboard](https://artificialanalysis.ai/evaluations/apex-agents-aa) |
| Gert Labs (Gert Labs Composite Game Benchmark) | 64.89 | Novel game environments | Agentic coding and decision-making | [Gert Labs rankings](https://gertlabs.com/rankings) |
| OSWorld-Verified | 75 | 369 real-world computer tasks (361 when eight Google Drive tasks are excluded) | Multi-step desktop and cross-application workflows | [OSWorld](https://os-world.github.io/) |
| CyberGym | 79.0 | 1,507 vulnerability analysis instances | Real-world cybersecurity | [CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale](https://www.cybergym.io/) |
| ExploitGym | 6.0 | 898 exploitation tasks | Advanced cybersecurity exploitation | [ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks?](https://arxiv.org/abs/2605.11086) |
| JobBench | 38.9 | 130 tasks across 35 occupations | Professional multi-source workflows | [JobBench: Aligning Agent Work With Human Will](https://arxiv.org/abs/2605.26329) |
| MCP Atlas | 70.6 | Tool-integrated agent tasks | Advanced tool use | [Introducing GPT-5.4 mini and nano](https://openai.com/index/introducing-gpt-5-4-mini-and-nano/) |
| Toolathlon | 54.6 | Multi-tool workflows | Advanced tool use | [Introducing GPT-5.4 mini and nano](https://openai.com/index/introducing-gpt-5-4-mini-and-nano/) |
| τ²-bench results (τ²-Bench Tool-Agent-User Evaluation) | 98.9 | Airline, retail, and telecom customer-service task sets | Dual-control customer-service workflows | [τ²-Bench: Evaluating Conversational Agents in a Dual-Control Environment](https://arxiv.org/abs/2506.07982) |
| DeepSearchQA | 73.6 | Agentic browsing and list-answer questions | Agentic web research | [Muse Spark Eval Methodology](https://ai.meta.com/static-resource/muse-spark-eval-methodology) |
| Claw-Eval | 60.3 | 300 tasks, 2,159 rubrics | Real-world general, multi-turn, and native multimodal agent execution | [Claw-Eval: Towards Trustworthy Evaluation of Autonomous Agents](https://arxiv.org/abs/2604.06132) |
| ResearchClawBench | 15.3 | 40 tasks across 10 scientific domains | Scientific research re-discovery | [ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research](https://arxiv.org/abs/2606.07591) |
| ApprenticeBench (ApprenticeBench: end-to-end computer use, continual learning, and long-horizon agency on a real accounts-payable job) | 11 | 100 vendor bills processed in sequence inside a simulated construction company | Long-horizon computer use with offline and online continual learning | [ApprenticeBench: a step change in AI's job readiness](https://neocognition.io/blog/apprentice-bench/) |

### Multimodal & Grounded

| Benchmark | Value | Tasks | Difficulty | Provenance |
| --- | --- | --- | --- | --- |
| MMMU-Pro (Massive Multi-discipline Multimodal Understanding Pro) | 81.2 | Multimodal academic reasoning | Frontier multimodal | [MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark](https://arxiv.org/abs/2409.02813) |
| AA-MMMU-Pro (Artificial Analysis MMMU-Pro) | 78.4 | Multimodal academic reasoning | Frontier multimodal | [Artificial Analysis MMMU-Pro Benchmark Leaderboard](https://artificialanalysis.ai/evaluations/mmmu-pro) |
| Design Arena Website (Design Arena Website Elo) | 1231 | Website generation comparisons | Design and website generation | [OpenRouter Grok 4.3 benchmarks](https://openrouter.ai/x-ai/grok-4.3/benchmarks) |
| OfficeQA Pro | 53.2 | Document and spreadsheet tasks | Enterprise grounded reasoning | [OfficeQA Pro: An Enterprise Benchmark for End-to-End Grounded Reasoning](https://arxiv.org/abs/2603.08655) |
| MMMU-Pro w/ Python (MMMU-Pro with Python) | 82.1 | Multimodal academic reasoning | Frontier multimodal | [Introducing GPT-5.4 mini and nano](https://openai.com/index/introducing-gpt-5-4-mini-and-nano/) |
| ERQA | 65.4 | Evidence-based visual QA | Grounded multimodal reasoning | [Qwen3.6 launch benchmarks](https://qwen.ai/blog?id=qwen3.6) |
| ScreenSpot Pro | 85.4 | 1,581 grounding instructions | Professional GUI grounding | [ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer Use](https://arxiv.org/abs/2504.07981) |
| MedXpertQA (MM) (MedXpertQA Multimodal) | 77.1 | 2,000 multimodal medical questions | Clinical multimodal reasoning | [Muse Spark Eval Methodology](https://ai.meta.com/static-resource/muse-spark-eval-methodology) |
| ZeroBench | 41.0 | 100 visual reasoning questions | Tool-augmented visual reasoning | [Muse Spark Eval Methodology](https://ai.meta.com/static-resource/muse-spark-eval-methodology) |
| SimpleVQA | 61.1 | Visual QA tasks | General visual understanding | [GLM-5V-Turbo](https://docs.z.ai/guides/vlm/glm-5v-turbo) |
| CharXiv (CharXiv Reasoning) | 82.8 | Scientific chart reasoning | Scientific visualization reasoning | [CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs](https://charxiv.github.io/) |

## Lifecycle and limitations log

Lifecycle events the source associates with this model.

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 claim

Values 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. Last attempted fetch for this model's score: 2026-09-22 10:17 UTC.