As of October 20, 2026, OpenAI GPT-5.4 ranks #30 of 213 models overall with 39.0 on the Artificial Analysis Intelligence Index. Its strongest category is reasoning, where it ranks #27 of 203. OpenAI GPT-5.4 costs $2.50 per million input tokens and $15.00 per million output tokens. It was released on March 5, 2026.
Its output price ranks 315 of 349 priced models, counting from the cheapest.
GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system. It features a 1M+ token context window (922K input, 128K output) with support for...
openai/gpt-5.4 The tick on a coding bar is the median of every model measured on it. Outside coding, the place is this model's rank on that board. A board that never ran the model says so.
Measured in 5 of 6 categories. A place counts only the models measured on that board.
The index averages only the coding benchmarks that have a result; the others are skipped, never filled in. How it is weighted
Not measured on SWE-bench Lite, SWE-bench (full test set), SWE-bench Multimodal, SWE-bench Multilingual, Terminal-Bench 4 (Vals AI), IOI (Vals AI), Terminal-Bench 2.1 (Vals AI), SWE-bench Multimodal (Vals AI), Vibe Code Bench 1-100 (Vals AI), CyberBench (Vals AI), LiveCodeBench (Artificial Analysis run), SciCode, FrontierCode, FrontierSWE, Cybench
Not measured on AA Agentic Index, Design Arena: agents building games, Design Arena: HTML slides, Design Arena: PowerPoint slides, Design Arena: spreadsheets, LiveBench research, LiveBench browser, LiveBench long agentic, SREBench (Vals AI), Finance Agent (Vals AI), Tax Agent Bench (Vals AI), Legal Research Bench (Vals AI), Harvey's Legal Agent Benchmark (Vals AI), Excel Modeling Benchmark (Vals AI), BioMysteryBench (Vals AI), Terminal-Bench-Science (Vals AI), Vals Science Index, Time Horizon Index (Vals AI), CUA-bench (Vals AI), Robo-bench (Vals AI), Remote Labor Index, GDPval, OSWorld 2.0, BALROG, TheAgentCompany, OSWorld, SpireClimb, LMArena Search
Not measured on LiveBench reasoning, MysteryMechanism (Vals AI), SimpleBench, VPCT, BIG-Bench Hard
Not measured on MATH 500 (Vals AI), MGSM (Vals AI), AA Math Index, AIME 2025 (Artificial Analysis run), MATH level 5, GSM8K
Not measured on MMLU Pro (Artificial Analysis run), MMLU, GeoBench, AI2 Reasoning Challenge
Not measured on LiveBench design
Not measured on Vals Index, LiveBench data analysis, Public Benefits Bench (Vals AI), Vals Multimodal Index, Fiction.LiveBench, Creative writing (Lech Mazur), Surface Evolver Bench, ExploitBench, HellaSwag, WinoGrande, PIQA, LMArena Vision: captioning
Not measured on Design Arena: images, Design Arena: graphic design, Design Arena: logos, LMArena Text to Image, LMArena Image Edit
Public boards shown as published (Vals AI, Artificial Analysis, Design Arena) and kept out of the coding index. Each links to its source on the benchmark page.
| Benchmark | Result | Index weight |
|---|---|---|
| Aider polyglotCoding | Not measured | 35% |
| SWE-bench VerifiedCoding | Not measured | 30% |
| SWE-bench LiteCoding | Not measured | Not in index |
| SWE-bench (full test set)Coding | Not measured | Not in index |
| SWE-bench MultimodalCoding | Not measured | Not in index |
| SWE-bench MultilingualCoding | Not measured | Not in index |
| LiveBench codingCoding | Not measured | 20% |
| LMArena WebDevCoding | 1464 | 15% |
| Vals IndexGeneral | Not measured | Not in index |
| GPQA DiamondReasoning | 91.7% | Not in index |
| AIMEMath | 96.7% | Not in index |
| MMLU ProKnowledge | 87.5% | Not in index |
| AA Intelligence IndexIntelligence | 39.0 | Not in index |
| AA Coding IndexCoding | 71.1 | Not in index |
| AA Agentic IndexAgentic work | Not measured | Not in index |
| Design Arena: websitesWeb design | 1199 | Not in index |
| Design Arena: UI componentsWeb design | 1216 | Not in index |
| Design Arena: game devWeb design | 1195 | Not in index |
| Design Arena: data vizWeb design | 1196 | Not in index |
| Design Arena: 3DWeb design | 1065 | Not in index |
| Design Arena: all code categoriesWeb design | 1180 | Not in index |
| Design Arena: SVGWeb design | 1141 | Not in index |
| Design Arena: ASCII artWeb design | 1189 | Not in index |
| Design Arena: agents building web appsAgentic work | 1059 | Not in index |
| Design Arena: agents building full-stack appsAgentic work | 1017 | Not in index |
| Design Arena: agents building mobile appsAgentic work | 1086 | Not in index |
| Design Arena: agents building Android appsAgentic work | 1047 | Not in index |
| Design Arena: agents building Godot gamesAgentic work | 1135 | Not in index |
| Design Arena: agents building gamesAgentic work | Not measured | Not in index |
| Design Arena: HTML slidesAgentic work | Not measured | Not in index |
| Design Arena: PowerPoint slidesAgentic work | Not measured | Not in index |
| Design Arena: spreadsheetsAgentic work | Not measured | Not in index |
| Design Arena: imagesImage generation | Not measured | Not in index |
| Design Arena: graphic designImage generation | Not measured | Not in index |
| Design Arena: logosImage generation | Not measured | Not in index |
| LiveBench reasoningReasoning | Not measured | Not in index |
| LiveBench data analysisGeneral | Not measured | Not in index |
| LiveBench researchAgentic work | Not measured | Not in index |
| LiveBench browserAgentic work | Not measured | Not in index |
| LiveBench long agenticAgentic work | Not measured | Not in index |
| LiveBench designWeb design | Not measured | Not in index |
| LiveCodeBench (Vals AI)Coding | 84.1% | Not in index |
| SWE-bench Verified (Vals AI)Coding | 78.2% | Not in index |
| Terminal-Bench 4 (Vals AI)Coding | Not measured | Not in index |
| Vibe Code Bench (Vals AI)Coding | 67.4% | Not in index |
| Code migration (Vals AI)Coding | 35.0% | Not in index |
| ProgramBench (Vals AI)Coding | 0.5% | Not in index |
| SREBench (Vals AI)Agentic work | Not measured | Not in index |
| IOI (Vals AI)Coding | Not measured | Not in index |
| Terminal-Bench 2.1 (Vals AI)Coding | Not measured | Not in index |
| SWE-bench Multimodal (Vals AI)Coding | Not measured | Not in index |
| Vibe Code Bench 1-100 (Vals AI)Coding | Not measured | Not in index |
| SkillsBench (Vals AI)Coding | 51.7% | Not in index |
| CyberBench (Vals AI)Coding | Not measured | Not in index |
| Finance Agent (Vals AI)Agentic work | Not measured | Not in index |
| Tax Agent Bench (Vals AI)Agentic work | Not measured | Not in index |
| Legal Research Bench (Vals AI)Agentic work | Not measured | Not in index |
| Harvey's Legal Agent Benchmark (Vals AI)Agentic work | Not measured | Not in index |
| Excel Modeling Benchmark (Vals AI)Agentic work | Not measured | Not in index |
| BioMysteryBench (Vals AI)Agentic work | Not measured | Not in index |
| Terminal-Bench-Science (Vals AI)Agentic work | Not measured | Not in index |
| Vals Science IndexAgentic work | Not measured | Not in index |
| Time Horizon Index (Vals AI)Agentic work | Not measured | Not in index |
| CUA-bench (Vals AI)Agentic work | Not measured | Not in index |
| Robo-bench (Vals AI)Agentic work | Not measured | Not in index |
| Public Benefits Bench (Vals AI)General | Not measured | Not in index |
| MysteryMechanism (Vals AI)Reasoning | Not measured | Not in index |
| MMMU Pro (Vals AI)Reasoning | 87.5% | Not in index |
| MATH 500 (Vals AI)Math | Not measured | Not in index |
| MGSM (Vals AI)Math | Not measured | Not in index |
| SAGE (Vals AI)General | 43.3% | Not in index |
| LegalBench (Vals AI)Knowledge | 86.0% | Not in index |
| CaseLaw (Vals AI)Knowledge | 63.8% | Not in index |
| CorpFin (Vals AI)Knowledge | 65.3% | Not in index |
| TaxEval (Vals AI)Knowledge | 74.0% | Not in index |
| MortgageTax (Vals AI)Knowledge | 68.3% | Not in index |
| MedQA (Vals AI)Knowledge | 96.1% | Not in index |
| MedCode (Vals AI)Knowledge | 41.3% | Not in index |
| MedScribe (Vals AI)General | 77.5% | Not in index |
| Vals Multimodal IndexGeneral | Not measured | Not in index |
| AA Math IndexMath | Not measured | Not in index |
| GPQA Diamond (Artificial Analysis run)Reasoning | 92.0% | Not in index |
| Humanity's Last Exam (Artificial Analysis run)Reasoning | 43.7% | Not in index |
| MMLU Pro (Artificial Analysis run)Knowledge | Not measured | Not in index |
| LiveCodeBench (Artificial Analysis run)Coding | Not measured | Not in index |
| SciCodeCoding | Not measured | Not in index |
| Terminal-Bench Hard (Artificial Analysis run)Coding | 57.6% | Not in index |
| IFBenchGeneral | 73.9% | Not in index |
| Long context reasoning (AA-LCR)General | 82.0% | Not in index |
| τ²-Bench TelecomAgentic work | 87.1% | Not in index |
| AIME 2025 (Artificial Analysis run)Math | Not measured | Not in index |
| Terminal-BenchCoding | 81.8% | Not in index |
| DeepSWECoding | 51.8% | Not in index |
| FrontierCodeCoding | Not measured | Not in index |
| FrontierSWECoding | Not measured | Not in index |
| GSO-BenchCoding | 31.4% | Not in index |
| WeirdMLCoding | 77.7% | Not in index |
| MirrorCodeCoding | 15.6% | Not in index |
| APEX-AgentsAgentic work | 52.4% | Not in index |
| METR time horizonsAgentic work | 74.3% | Not in index |
| Remote Labor IndexAgentic work | Not measured | Not in index |
| GDPvalAgentic work | Not measured | Not in index |
| OSWorld 2.0Agentic work | Not measured | Not in index |
| BALROGAgentic work | Not measured | Not in index |
| DeepResearch BenchAgentic work | 35.1% | Not in index |
| PostTrainBenchAgentic work | 19.0% | Not in index |
| GPQA Diamond (Epoch AI run)Reasoning | 93.3% | Not in index |
| ARC-AGIReasoning | 93.7% | Not in index |
| ARC-AGI-2Reasoning | 74.0% | Not in index |
| Humanity's Last ExamReasoning | 36.2% | Not in index |
| SimpleBenchReasoning | Not measured | Not in index |
| Chess puzzlesReasoning | 48.0% | Not in index |
| Mystery game puzzlesReasoning | 37.0% | Not in index |
| EBR-benchReasoning | 25.4% | Not in index |
| OTIS Mock AIMEMath | 97.8% | Not in index |
| FrontierMath, tiers 1 to 3Math | 78.6% | Not in index |
| FrontierMath, tier 4Math | 49.0% | Not in index |
| ProofBenchMath | 56.0% | Not in index |
| MATH level 5Math | Not measured | Not in index |
| GSM8KMath | Not measured | Not in index |
| SimpleQA VerifiedKnowledge | 45.1% | Not in index |
| MMLUKnowledge | Not measured | Not in index |
| GeoBenchKnowledge | Not measured | Not in index |
| DTBenchGeneral | 94.4% | Not in index |
| LMCAGeneral | 52.0% | Not in index |
| Fiction.LiveBenchGeneral | Not measured | Not in index |
| Creative writing (Lech Mazur)General | Not measured | Not in index |
| CL-benchGeneral | 27.9% | Not in index |
| Surface Evolver BenchGeneral | Not measured | Not in index |
| Furniture assemblyGeneral | 37.5% | Not in index |
| ExploitBenchGeneral | Not measured | Not in index |
| CybenchCoding | Not measured | Not in index |
| TheAgentCompanyAgentic work | Not measured | Not in index |
| OSWorldAgentic work | Not measured | Not in index |
| SpireClimbAgentic work | Not measured | Not in index |
| VPCTReasoning | Not measured | Not in index |
| BIG-Bench HardReasoning | Not measured | Not in index |
| AI2 Reasoning ChallengeKnowledge | Not measured | Not in index |
| HellaSwagGeneral | Not measured | Not in index |
| WinoGrandeGeneral | Not measured | Not in index |
| PIQAGeneral | Not measured | Not in index |
| CL-bench LifeGeneral | 21.7% | Not in index |
| ALE-BenchCoding | 1607.0 | Not in index |
| Vending-Bench 2Agentic work | $6,144 | Not in index |
| Epoch Capabilities IndexIntelligence | 156.7 | Not in index |
| LMArena TextGeneral | 1469 | Not in index |
| LMArena Text: codingCoding | 1496 | Not in index |
| LMArena Text: hard promptsReasoning | 1485 | Not in index |
| LMArena Text: mathMath | 1487 | Not in index |
| LMArena Text: instruction followingGeneral | 1469 | Not in index |
| LMArena Text: creative writingGeneral | 1439 | Not in index |
| LMArena VisionGeneral | 1303 | Not in index |
| LMArena Text: EnglishGeneral | 1467 | Not in index |
| LMArena Text: non-EnglishGeneral | 1465 | Not in index |
| LMArena Text: ChineseGeneral | 1518 | Not in index |
| LMArena Text: GermanGeneral | 1473 | Not in index |
| LMArena Text: JapaneseGeneral | 1485 | Not in index |
| LMArena Text: KoreanGeneral | 1448 | Not in index |
| LMArena Text: FrenchGeneral | 1493 | Not in index |
| LMArena Text: SpanishGeneral | 1453 | Not in index |
| LMArena Text: RussianGeneral | 1480 | Not in index |
| LMArena Text: PolishGeneral | 1472 | Not in index |
| LMArena Text: multi-turnGeneral | 1482 | Not in index |
| LMArena Text: longer queriesGeneral | 1473 | Not in index |
| LMArena Text: expertReasoning | 1506 | Not in index |
| LMArena Text: hard prompts in EnglishReasoning | 1479 | Not in index |
| LMArena Text: software and ITCoding | 1491 | Not in index |
| LMArena Text: mathematical workMath | 1497 | Not in index |
| LMArena Text: scienceKnowledge | 1478 | Not in index |
| LMArena Text: medicine and healthcareKnowledge | 1458 | Not in index |
| LMArena Text: legal and governmentKnowledge | 1486 | Not in index |
| LMArena Text: business and financeKnowledge | 1474 | Not in index |
| LMArena Text: writing and languageGeneral | 1458 | Not in index |
| LMArena Text: entertainment and mediaGeneral | 1438 | Not in index |
| LMArena WebDev: ReactCoding | 1454 | Not in index |
| LMArena WebDev: HTMLCoding | 1478 | Not in index |
| LMArena WebDev: from an imageCoding | 1436 | Not in index |
| LMArena Vision: EnglishGeneral | 1301 | Not in index |
| LMArena Vision: ChineseGeneral | 1354 | Not in index |
| LMArena Vision: reading textGeneral | 1312 | Not in index |
| LMArena Vision: diagramsReasoning | 1327 | Not in index |
| LMArena Vision: homeworkReasoning | 1336 | Not in index |
| LMArena Vision: humorGeneral | 1284 | Not in index |
| LMArena Vision: captioningGeneral | Not measured | Not in index |
| LMArena Vision: recognizing thingsGeneral | 1251 | Not in index |
| LMArena Vision: creative writingGeneral | 1283 | Not in index |
| LMArena SearchAgentic work | Not measured | Not in index |
| LMArena DocumentGeneral | 1471 | Not in index |
| LMArena Text to ImageImage generation | Not measured | Not in index |
| LMArena Image EditImage generation | Not measured | Not in index |
A list price says what a token costs. The cost per solved task says what a working answer costs, which is the number that decides.
Beside a result, many boards publish what the run cost, how long it took, the tokens it spent or the votes behind a rating. Each figure as its board printed it, one board per row.
| Board | Result | What the run cost | Also published |
|---|---|---|---|
| SWE-bench Verified (Vals AI) | 78.2% | $0.80 per test | Duration, as printed 5m 07s |
| Vibe Code Bench (Vals AI) | 67.4% | $10.69 per test | Duration, as printed 1h 26m |
| Code migration (Vals AI) | 35.0% | $7.76 per test | Duration, as printed 35m 46s |
| ProgramBench (Vals AI) | 0.5% | $2.27 per test | Duration, as printed 15m 58s |
| SkillsBench (Vals AI) | 51.7% | $2.04 per test | Duration, as printed 9m 02s |
| DeepSWE | 51.8% | $5.6525 mean cost per task | Solved in 4 tries 77.9% · Output tokens per task 71,409 · Agent steps per task 70 |
| WeirdML | 77.7% | $5.7199 per run | Median lines of code 968 |
| ARC-AGI | 93.7% | $0.62 per task | Nothing else |
| ARC-AGI-2 | 74.0% | $1.52 per task | Nothing else |
| ProofBench | 56.0% | $3.202 per test | Time per proof 45m 52s |
| ALE-Bench | 1607.0 | $0.6122 per problem | Tokens per problem 45,461 |
| LMArena WebDev | 1464 Elo | Not published | Votes 1,350 · Interval (±) ±18.9 |
| GPQA Diamond | 91.7% | Not published | Latency, as printed 3m 23s |
|---|---|---|---|
| AIME | 96.7% | Not published | Latency, as printed 2m 01s |
| MMLU Pro | 87.5% | Not published | Latency, as printed 29 s |
| Design Arena: websites | 1199 Elo | Not published | Battles 13,180 |
| Design Arena: UI components | 1216 Elo | Not published | Battles 129 |
| Design Arena: game dev | 1195 Elo | Not published | Battles 4,000 |
| Design Arena: data viz | 1196 Elo | Not published | Battles 1,191 |
| Design Arena: 3D | 1065 Elo | Not published | Battles 4,300 |
| Design Arena: all code categories | 1180 Elo | Not published | Battles 21,152 |
| Design Arena: SVG | 1141 Elo | Not published | Battles 3,842 |
| Design Arena: ASCII art | 1189 Elo | Not published | Battles 1,310 |
| Design Arena: agents building web apps | 1059 Elo | Not published | Battles 3,319 |
| Design Arena: agents building full-stack apps | 1017 Elo | Not published | Battles 2,152 |
| Design Arena: agents building mobile apps | 1086 Elo | Not published | Battles 1,929 |
| Design Arena: agents building Android apps | 1047 Elo | Not published | Battles 502 |
| LiveCodeBench (Vals AI) | 84.1% | Not published | Latency, as printed 3m 29s |
| MMMU Pro (Vals AI) | 87.5% | Not published | Latency, as printed 1m 06s |
| SAGE (Vals AI) | 43.3% | Not published | Latency, as printed 2m 32s |
| LegalBench (Vals AI) | 86.0% | Not published | Latency, as printed 28 s |
| CaseLaw (Vals AI) | 63.8% | Not published | Latency, as printed 57 s |
| CorpFin (Vals AI) | 65.3% | Not published | Latency, as printed 1m 22s |
| TaxEval (Vals AI) | 74.0% | Not published | Latency, as printed 1m 19s |
| MortgageTax (Vals AI) | 68.3% | Not published | Latency, as printed 45 s |
| MedQA (Vals AI) | 96.1% | Not published | Latency, as printed 23 s |
| MedCode (Vals AI) | 41.3% | Not published | Latency, as printed 3m 07s |
| MedScribe (Vals AI) | 77.5% | Not published | Latency, as printed 4m 43s |
| APEX-Agents | 52.4% | Not published | Mean score 68.2% |
| LMArena Text | 1469 Elo | Not published | Votes 64,776 · Interval (±) ±3.7 |
| LMArena Text: coding | 1496 Elo | Not published | Votes 17,600 · Interval (±) ±5.8 |
| LMArena Text: hard prompts | 1485 Elo | Not published | Votes 42,140 · Interval (±) ±4.5 |
| LMArena Text: math | 1487 Elo | Not published | Votes 3,379 · Interval (±) ±10.6 |
| LMArena Text: instruction following | 1469 Elo | Not published | Votes 21,877 · Interval (±) ±5.4 |
| LMArena Text: creative writing | 1439 Elo | Not published | Votes 11,128 · Interval (±) ±6.9 |
| LMArena Vision | 1303 Elo | Not published | Votes 30,747 · Interval (±) ±5.7 |
| LMArena Text: English | 1467 Elo | Not published | Votes 29,330 · Interval (±) ±4.9 |
| LMArena Text: non-English | 1465 Elo | Not published | Votes 35,444 · Interval (±) ±4.6 |
| LMArena Text: Chinese | 1518 Elo | Not published | Votes 4,141 · Interval (±) ±9.9 |
| LMArena Text: German | 1473 Elo | Not published | Votes 1,095 · Interval (±) ±18.6 |
| LMArena Text: Japanese | 1485 Elo | Not published | Votes 676 · Interval (±) ±24.7 |
| LMArena Text: Korean | 1448 Elo | Not published | Votes 1,112 · Interval (±) ±19.5 |
| LMArena Text: French | 1493 Elo | Not published | Votes 2,395 · Interval (±) ±13.9 |
| LMArena Text: Spanish | 1453 Elo | Not published | Votes 2,060 · Interval (±) ±14.1 |
| LMArena Text: Russian | 1480 Elo | Not published | Votes 7,202 · Interval (±) ±7.8 |
| LMArena Text: Polish | 1472 Elo | Not published | Votes 1,318 · Interval (±) ±16.6 |
| LMArena Text: multi-turn | 1482 Elo | Not published | Votes 11,988 · Interval (±) ±6.6 |
| LMArena Text: longer queries | 1473 Elo | Not published | Votes 28,438 · Interval (±) ±5.2 |
| LMArena Text: expert | 1506 Elo | Not published | Votes 6,359 · Interval (±) ±8.3 |
| LMArena Text: hard prompts in English | 1479 Elo | Not published | Votes 19,722 · Interval (±) ±5.6 |
| LMArena Text: software and IT | 1491 Elo | Not published | Votes 25,473 · Interval (±) ±5.1 |
| LMArena Text: mathematical work | 1497 Elo | Not published | Votes 3,544 · Interval (±) ±10.6 |
| LMArena Text: science | 1478 Elo | Not published | Votes 10,648 · Interval (±) ±6.7 |
| LMArena Text: medicine and healthcare | 1458 Elo | Not published | Votes 4,839 · Interval (±) ±9.3 |
| LMArena Text: legal and government | 1486 Elo | Not published | Votes 5,239 · Interval (±) ±9.0 |
| LMArena Text: business and finance | 1474 Elo | Not published | Votes 13,095 · Interval (±) ±6.4 |
| LMArena Text: writing and language | 1458 Elo | Not published | Votes 16,350 · Interval (±) ±5.9 |
| LMArena Text: entertainment and media | 1438 Elo | Not published | Votes 14,154 · Interval (±) ±6.3 |
| LMArena WebDev: React | 1454 Elo | Not published | Votes 1,234 · Interval (±) ±19.8 |
| LMArena WebDev: HTML | 1478 Elo | Not published | Votes 132 · Interval (±) ±52.9 |
| LMArena WebDev: from an image | 1436 Elo | Not published | Votes 3,201 · Interval (±) ±11.7 |
| LMArena Vision: English | 1301 Elo | Not published | Votes 12,132 · Interval (±) ±7.9 |
| LMArena Vision: Chinese | 1354 Elo | Not published | Votes 1,846 · Interval (±) ±16.2 |
| LMArena Vision: reading text | 1312 Elo | Not published | Votes 21,790 · Interval (±) ±5.6 |
| LMArena Vision: diagrams | 1327 Elo | Not published | Votes 8,230 · Interval (±) ±8.0 |
| LMArena Vision: homework | 1336 Elo | Not published | Votes 4,003 · Interval (±) ±10.3 |
| LMArena Vision: humor | 1284 Elo | Not published | Votes 1,143 · Interval (±) ±18.3 |
| LMArena Vision: recognizing things | 1251 Elo | Not published | Votes 137 · Interval (±) ±45.9 |
| LMArena Vision: creative writing | 1283 Elo | Not published | Votes 2,112 · Interval (±) ±13.8 |
| LMArena Document | 1471 Elo | Not published | Votes 33,331 · Interval (±) ±6.0 |