As of October 20, 2026, OpenAI GPT-5.2 ranks #48 of 213 models overall with 30.4 on the Artificial Analysis Intelligence Index. Its strongest category is reasoning, where it ranks #27 of 203. OpenAI GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens. It was released on December 10, 2025.
Its output price ranks 311 of 349 priced models, counting from the cheapest.
GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1. It uses adaptive reasoning to allocate computation dynamically, responding quickly...
openai/gpt-5.2 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, AA Coding Index, Terminal-Bench 4 (Vals AI), Code migration (Vals AI), ProgramBench (Vals AI), IOI (Vals AI), Terminal-Bench 2.1 (Vals AI), SWE-bench Multimodal (Vals AI), Vibe Code Bench 1-100 (Vals AI), SkillsBench (Vals AI), CyberBench (Vals AI), SciCode, DeepSWE, FrontierCode, FrontierSWE, MirrorCode, Cybench, LMArena WebDev: React, LMArena WebDev: from an image
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), APEX-Agents, OSWorld 2.0, BALROG, PostTrainBench, TheAgentCompany, OSWorld, SpireClimb, LMArena Search
Not measured on LiveBench reasoning, MysteryMechanism (Vals AI), BIG-Bench Hard
Not measured on MATH 500 (Vals AI), MATH level 5, GSM8K
Not measured on 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, CL-bench Life
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 | 72.8% | 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 | 66.7% | Not in index |
| LiveBench codingCoding | Not measured | 20% |
| LMArena WebDevCoding | 1417 | 15% |
| Vals IndexGeneral | Not measured | Not in index |
| GPQA DiamondReasoning | 91.7% | Not in index |
| AIMEMath | 96.9% | Not in index |
| MMLU ProKnowledge | 86.2% | Not in index |
| AA Intelligence IndexIntelligence | 30.4 | Not in index |
| AA Coding IndexCoding | Not measured | Not in index |
| AA Agentic IndexAgentic work | Not measured | Not in index |
| Design Arena: websitesWeb design | 1196 | Not in index |
| Design Arena: UI componentsWeb design | 1190 | Not in index |
| Design Arena: game devWeb design | 1196 | Not in index |
| Design Arena: data vizWeb design | 1202 | Not in index |
| Design Arena: 3DWeb design | 1079 | Not in index |
| Design Arena: all code categoriesWeb design | 1171 | Not in index |
| Design Arena: SVGWeb design | 1137 | Not in index |
| Design Arena: ASCII artWeb design | 1153 | Not in index |
| Design Arena: agents building web appsAgentic work | 1086 | Not in index |
| Design Arena: agents building full-stack appsAgentic work | 1043 | Not in index |
| Design Arena: agents building mobile appsAgentic work | 1093 | Not in index |
| Design Arena: agents building Android appsAgentic work | 1075 | Not in index |
| Design Arena: agents building Godot gamesAgentic work | 1142 | 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 | 85.4% | Not in index |
| SWE-bench Verified (Vals AI)Coding | 75.8% | Not in index |
| Terminal-Bench 4 (Vals AI)Coding | Not measured | Not in index |
| Vibe Code Bench (Vals AI)Coding | 53.5% | Not in index |
| Code migration (Vals AI)Coding | Not measured | Not in index |
| ProgramBench (Vals AI)Coding | Not measured | 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 | Not measured | 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 | 86.7% | Not in index |
| MATH 500 (Vals AI)Math | Not measured | Not in index |
| MGSM (Vals AI)Math | 94.0% | Not in index |
| SAGE (Vals AI)General | 49.3% | Not in index |
| LegalBench (Vals AI)Knowledge | 82.8% | Not in index |
| CaseLaw (Vals AI)Knowledge | 66.0% | Not in index |
| CorpFin (Vals AI)Knowledge | 65.9% | Not in index |
| TaxEval (Vals AI)Knowledge | 75.8% | Not in index |
| MortgageTax (Vals AI)Knowledge | 67.1% | Not in index |
| MedQA (Vals AI)Knowledge | 94.1% | Not in index |
| MedCode (Vals AI)Knowledge | 49.8% | Not in index |
| MedScribe (Vals AI)General | 84.4% | Not in index |
| Vals Multimodal IndexGeneral | Not measured | Not in index |
| AA Math IndexMath | 99.0 | Not in index |
| GPQA Diamond (Artificial Analysis run)Reasoning | 90.3% | Not in index |
| Humanity's Last Exam (Artificial Analysis run)Reasoning | 37.7% | Not in index |
| MMLU Pro (Artificial Analysis run)Knowledge | 87.4% | Not in index |
| LiveCodeBench (Artificial Analysis run)Coding | 89.4% | Not in index |
| SciCodeCoding | Not measured | Not in index |
| Terminal-Bench Hard (Artificial Analysis run)Coding | 47.0% | Not in index |
| IFBenchGeneral | 75.4% | Not in index |
| Long context reasoning (AA-LCR)General | 82.7% | Not in index |
| τ²-Bench TelecomAgentic work | 84.8% | Not in index |
| AIME 2025 (Artificial Analysis run)Math | 99.0% | Not in index |
| Terminal-BenchCoding | 64.9% | Not in index |
| DeepSWECoding | Not measured | Not in index |
| FrontierCodeCoding | Not measured | Not in index |
| FrontierSWECoding | Not measured | Not in index |
| GSO-BenchCoding | 27.5% | Not in index |
| WeirdMLCoding | 72.2% | Not in index |
| MirrorCodeCoding | Not measured | Not in index |
| APEX-AgentsAgentic work | Not measured | Not in index |
| METR time horizonsAgentic work | 75.3% | Not in index |
| Remote Labor IndexAgentic work | 2.5% | Not in index |
| GDPvalAgentic work | 49.7% | Not in index |
| OSWorld 2.0Agentic work | Not measured | Not in index |
| BALROGAgentic work | Not measured | Not in index |
| DeepResearch BenchAgentic work | 41.1% | Not in index |
| PostTrainBenchAgentic work | Not measured | Not in index |
| GPQA Diamond (Epoch AI run)Reasoning | 91.4% | Not in index |
| ARC-AGIReasoning | 86.2% | Not in index |
| ARC-AGI-2Reasoning | 52.9% | Not in index |
| Humanity's Last ExamReasoning | 27.8% | Not in index |
| SimpleBenchReasoning | 45.8% | Not in index |
| Chess puzzlesReasoning | 48.0% | Not in index |
| Mystery game puzzlesReasoning | 23.0% | Not in index |
| EBR-benchReasoning | 23.0% | Not in index |
| OTIS Mock AIMEMath | 96.1% | Not in index |
| FrontierMath, tiers 1 to 3Math | 67.4% | Not in index |
| FrontierMath, tier 4Math | 31.7% | Not in index |
| ProofBenchMath | 15.0% | Not in index |
| MATH level 5Math | Not measured | Not in index |
| GSM8KMath | Not measured | Not in index |
| SimpleQA VerifiedKnowledge | 37.1% | Not in index |
| MMLUKnowledge | Not measured | Not in index |
| GeoBenchKnowledge | Not measured | Not in index |
| DTBenchGeneral | 90.9% | Not in index |
| LMCAGeneral | 43.9% | Not in index |
| Fiction.LiveBenchGeneral | Not measured | Not in index |
| Creative writing (Lech Mazur)General | Not measured | Not in index |
| CL-benchGeneral | 18.2% | Not in index |
| Surface Evolver BenchGeneral | Not measured | Not in index |
| Furniture assemblyGeneral | 38.3% | 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 | 84.0% | 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 | Not measured | Not in index |
| ALE-BenchCoding | 1293.5 | Not in index |
| Vending-Bench 2Agentic work | $3,591 | Not in index |
| Epoch Capabilities IndexIntelligence | 153.5 | Not in index |
| LMArena TextGeneral | 1416 | Not in index |
| LMArena Text: codingCoding | 1442 | Not in index |
| LMArena Text: hard promptsReasoning | 1428 | Not in index |
| LMArena Text: mathMath | 1440 | Not in index |
| LMArena Text: instruction followingGeneral | 1409 | Not in index |
| LMArena Text: creative writingGeneral | 1376 | Not in index |
| LMArena VisionGeneral | 1246 | Not in index |
| LMArena Text: EnglishGeneral | 1422 | Not in index |
| LMArena Text: non-EnglishGeneral | 1405 | Not in index |
| LMArena Text: ChineseGeneral | 1447 | Not in index |
| LMArena Text: GermanGeneral | 1444 | Not in index |
| LMArena Text: JapaneseGeneral | 1420 | Not in index |
| LMArena Text: KoreanGeneral | 1374 | Not in index |
| LMArena Text: FrenchGeneral | 1454 | Not in index |
| LMArena Text: SpanishGeneral | 1413 | Not in index |
| LMArena Text: RussianGeneral | 1415 | Not in index |
| LMArena Text: PolishGeneral | 1426 | Not in index |
| LMArena Text: multi-turnGeneral | 1419 | Not in index |
| LMArena Text: longer queriesGeneral | 1413 | Not in index |
| LMArena Text: expertReasoning | 1445 | Not in index |
| LMArena Text: hard prompts in EnglishReasoning | 1431 | Not in index |
| LMArena Text: software and ITCoding | 1434 | Not in index |
| LMArena Text: mathematical workMath | 1442 | Not in index |
| LMArena Text: scienceKnowledge | 1425 | Not in index |
| LMArena Text: medicine and healthcareKnowledge | 1425 | Not in index |
| LMArena Text: legal and governmentKnowledge | 1427 | Not in index |
| LMArena Text: business and financeKnowledge | 1414 | Not in index |
| LMArena Text: writing and languageGeneral | 1392 | Not in index |
| LMArena Text: entertainment and mediaGeneral | 1378 | Not in index |
| LMArena WebDev: ReactCoding | Not measured | Not in index |
| LMArena WebDev: HTMLCoding | 1427 | Not in index |
| LMArena WebDev: from an imageCoding | Not measured | Not in index |
| LMArena Vision: EnglishGeneral | 1243 | Not in index |
| LMArena Vision: ChineseGeneral | 1305 | Not in index |
| LMArena Vision: reading textGeneral | 1257 | Not in index |
| LMArena Vision: diagramsReasoning | 1265 | Not in index |
| LMArena Vision: homeworkReasoning | 1276 | Not in index |
| LMArena Vision: humorGeneral | 1227 | Not in index |
| LMArena Vision: captioningGeneral | 1267 | Not in index |
| LMArena Vision: recognizing thingsGeneral | 1232 | Not in index |
| LMArena Vision: creative writingGeneral | 1225 | Not in index |
| LMArena SearchAgentic work | Not measured | Not in index |
| LMArena DocumentGeneral | 1405 | 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 | 72.8% | $0.4736 per attempt | Nothing else |
| SWE-bench Multilingual | 66.7% | $0.5364 per attempt | Nothing else |
| SWE-bench Verified (Vals AI) | 75.8% | $1.45 per test | Duration, as printed 10m 40s |
| Vibe Code Bench (Vals AI) | 53.5% | $17.75 per test | Duration, as printed 1h 23m |
| WeirdML | 72.2% | $2.0474 per run | Median lines of code 394 |
| ARC-AGI | 86.2% | $0.96 per task | Nothing else |
| ARC-AGI-2 | 52.9% | $1.8982 per task | Nothing else |
| ProofBench | 15.0% | $1.0597 per test | Time per proof 24m 54s |
| ALE-Bench | 1293.5 | $0.2969 per problem | Tokens per problem 24,031 |
| LMArena WebDev | 1417 Elo | Not published | Votes 1,081 · Interval (±) ±22.6 |
| GPQA Diamond | 91.7% | Not published | Latency, as printed 2m 53s |
| AIME | 96.9% | Not published | Latency, as printed 2m 00s |
| MMLU Pro | 86.2% | Not published | Latency, as printed 35 s |
|---|---|---|---|
| Design Arena: agents building web apps | 1086 Elo | Not published | Battles 3,570 |
| Design Arena: agents building full-stack apps | 1043 Elo | Not published | Battles 2,165 |
| Design Arena: agents building mobile apps | 1093 Elo | Not published | Battles 1,813 |
| Design Arena: agents building Android apps | 1075 Elo | Not published | Battles 660 |
| LiveCodeBench (Vals AI) | 85.4% | Not published | Latency, as printed 2m 32s |
| MMMU Pro (Vals AI) | 86.7% | Not published | Latency, as printed 1m 36s |
| MGSM (Vals AI) | 94.0% | Not published | Latency, as printed 11 s |
| SAGE (Vals AI) | 49.3% | Not published | Latency, as printed 2m 59s |
| LegalBench (Vals AI) | 82.8% | Not published | Latency, as printed 17 s |
| CaseLaw (Vals AI) | 66.0% | Not published | Latency, as printed 1m 26s |
| CorpFin (Vals AI) | 65.9% | Not published | Latency, as printed 26 s |
| TaxEval (Vals AI) | 75.8% | Not published | Latency, as printed 1m 00s |
| MortgageTax (Vals AI) | 67.1% | Not published | Latency, as printed 1m 11s |
| MedQA (Vals AI) | 94.1% | Not published | Latency, as printed 19 s |
| MedCode (Vals AI) | 49.8% | Not published | Latency, as printed 2m 37s |
| MedScribe (Vals AI) | 84.4% | Not published | Latency, as printed 2m 06s |
| LMArena Text | 1416 Elo | Not published | Votes 49,655 · Interval (±) ±3.5 |
| LMArena Text: coding | 1442 Elo | Not published | Votes 12,245 · Interval (±) ±6.0 |
| LMArena Text: hard prompts | 1428 Elo | Not published | Votes 29,889 · Interval (±) ±4.4 |
| LMArena Text: math | 1440 Elo | Not published | Votes 3,127 · Interval (±) ±10.7 |
| LMArena Text: instruction following | 1409 Elo | Not published | Votes 15,300 · Interval (±) ±5.5 |
| LMArena Text: creative writing | 1376 Elo | Not published | Votes 7,660 · Interval (±) ±7.3 |
| LMArena Vision | 1246 Elo | Not published | Votes 19,022 · Interval (±) ±6.0 |
| LMArena Text: English | 1422 Elo | Not published | Votes 21,224 · Interval (±) ±4.9 |
| LMArena Text: non-English | 1405 Elo | Not published | Votes 46,249 · Interval (±) ±3.9 |
| LMArena Text: Chinese | 1447 Elo | Not published | Votes 3,575 · Interval (±) ±10.2 |
| LMArena Text: German | 1444 Elo | Not published | Votes 834 · Interval (±) ±20.8 |
| LMArena Text: Japanese | 1420 Elo | Not published | Votes 446 · Interval (±) ±29.5 |
| LMArena Text: Korean | 1374 Elo | Not published | Votes 1,302 · Interval (±) ±17.3 |
| LMArena Text: French | 1454 Elo | Not published | Votes 1,440 · Interval (±) ±16.9 |
| LMArena Text: Spanish | 1413 Elo | Not published | Votes 1,536 · Interval (±) ±15.6 |
| LMArena Text: Russian | 1415 Elo | Not published | Votes 9,726 · Interval (±) ±6.6 |
| LMArena Text: Polish | 1426 Elo | Not published | Votes 1,631 · Interval (±) ±14.9 |
| LMArena Text: multi-turn | 1419 Elo | Not published | Votes 9,245 · Interval (±) ±6.8 |
| LMArena Text: longer queries | 1413 Elo | Not published | Votes 34,073 · Interval (±) ±4.7 |
| LMArena Text: expert | 1445 Elo | Not published | Votes 4,001 · Interval (±) ±9.6 |
| LMArena Text: hard prompts in English | 1431 Elo | Not published | Votes 13,040 · Interval (±) ±5.9 |
| LMArena Text: software and IT | 1434 Elo | Not published | Votes 18,421 · Interval (±) ±5.1 |
| LMArena Text: mathematical work | 1442 Elo | Not published | Votes 2,622 · Interval (±) ±11.9 |
| LMArena Text: science | 1425 Elo | Not published | Votes 8,275 · Interval (±) ±7.0 |
| LMArena Text: medicine and healthcare | 1425 Elo | Not published | Votes 6,179 · Interval (±) ±8.3 |
| LMArena Text: legal and government | 1427 Elo | Not published | Votes 3,769 · Interval (±) ±9.9 |
| LMArena Text: business and finance | 1414 Elo | Not published | Votes 9,758 · Interval (±) ±6.5 |
| LMArena Text: writing and language | 1392 Elo | Not published | Votes 11,510 · Interval (±) ±6.1 |
| LMArena Text: entertainment and media | 1378 Elo | Not published | Votes 17,744 · Interval (±) ±5.5 |
| LMArena WebDev: HTML | 1427 Elo | Not published | Votes 1,081 · Interval (±) ±23.5 |
| LMArena Vision: English | 1243 Elo | Not published | Votes 7,347 · Interval (±) ±8.7 |
| LMArena Vision: Chinese | 1305 Elo | Not published | Votes 1,351 · Interval (±) ±17.9 |
| LMArena Vision: reading text | 1257 Elo | Not published | Votes 12,905 · Interval (±) ±6.4 |
| LMArena Vision: diagrams | 1265 Elo | Not published | Votes 4,773 · Interval (±) ±9.5 |
| LMArena Vision: homework | 1276 Elo | Not published | Votes 2,888 · Interval (±) ±11.6 |
| LMArena Vision: humor | 1227 Elo | Not published | Votes 681 · Interval (±) ±23.6 |
| LMArena Vision: captioning | 1267 Elo | Not published | Votes 94 · Interval (±) ±58.0 |
| LMArena Vision: recognizing things | 1232 Elo | Not published | Votes 168 · Interval (±) ±41.1 |
| LMArena Vision: creative writing | 1225 Elo | Not published | Votes 1,290 · Interval (±) ±17.3 |
| LMArena Document | 1405 Elo | Not published | Votes 7,108 · Interval (±) ±9.5 |