# OpenAI GPT-5.2 As of October 22, 2026, OpenAI GPT-5.2 ranks #40 of 123 models overall with 60.7 ±7.7 on the LLMBenchmark Index. GPT-5.2 ranks #35 of 108 in math, its strongest category. OpenAI GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens. GPT-5.2 was released on December 10, 2025. Page: https://llmbenchmark.ai/models/gpt-5-2 Data from October 22, 2026. Every figure below names the board it was read from and the day. - Maker: OpenAI - Released: December 10, 2025 - Price per million input tokens: $1.75 - Price per million output tokens: $14.00 - Context window: 400K tokens - Overall rank: 40 of 123, by the LLMBenchmark Index - LLMBenchmark Index: 60.7 ±7.7 (90% margin, from 6 of 7 inputs) - Artificial Analysis Intelligence Index: 30.4 - Vibe Coding Index: 55.9 (from 3 of 6 coding benchmarks) Benchmark results (114 of 195): - SWE-bench Verified: 72.8% (source: SWE-bench (swebench.com), read October 22, 2026) - SWE-bench Multilingual: 66.7% (source: SWE-bench (swebench.com), read October 22, 2026) - LMArena WebDev: 1417 Elo (source: LMArena WebDev, read October 22, 2026) - GPQA Diamond: 91.7% (source: GPQA (Vals AI), read October 22, 2026) - AIME: 96.9% (source: AIME (Vals AI), read October 22, 2026) - MMLU Pro: 86.2% (source: MMLU Pro (Vals AI), read October 22, 2026) - AA Intelligence Index: 30.4 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - Design Arena: websites: 1195 Elo (source: Design Arena, read October 22, 2026) - Design Arena: UI components: 1190 Elo (source: Design Arena, read October 22, 2026) - Design Arena: game dev: 1195 Elo (source: Design Arena, read October 22, 2026) - Design Arena: data viz: 1202 Elo (source: Design Arena, read October 22, 2026) - Design Arena: 3D: 1079 Elo (source: Design Arena, read October 22, 2026) - Design Arena: all code categories: 1170 Elo (source: Design Arena, read October 22, 2026) - Design Arena: SVG: 1139 Elo (source: Design Arena, read October 22, 2026) - Design Arena: ASCII art: 1152 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building web apps: 1086 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building full-stack apps: 1043 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building mobile apps: 1093 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building Android apps: 1059 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building Godot games: 1142 Elo (source: Design Arena, read October 22, 2026) - LiveCodeBench (Vals AI): 85.4% (source: Vals AI boards, read October 22, 2026) - SWE-bench Verified (Vals AI): 75.8% (source: Vals AI boards, read October 22, 2026) - Vibe Code Bench (Vals AI): 53.5% (source: Vals AI boards, read October 22, 2026) - MMMU Pro (Vals AI): 86.7% (source: Vals AI boards, read October 22, 2026) - MGSM (Vals AI): 94.0% (source: Vals AI boards, read October 22, 2026) - SAGE (Vals AI): 49.3% (source: Vals AI boards, read October 22, 2026) - LegalBench (Vals AI): 82.8% (source: Vals AI boards, read October 22, 2026) - CaseLaw (Vals AI): 66.0% (source: Vals AI boards, read October 22, 2026) - CorpFin (Vals AI): 65.9% (source: Vals AI boards, read October 22, 2026) - TaxEval (Vals AI): 75.8% (source: Vals AI boards, read October 22, 2026) - MortgageTax (Vals AI): 67.1% (source: Vals AI boards, read October 22, 2026) - MedQA (Vals AI): 94.1% (source: Vals AI boards, read October 22, 2026) - MedCode (Vals AI): 49.8% (source: Vals AI boards, read October 22, 2026) - MedScribe (Vals AI): 84.4% (source: Vals AI boards, read October 22, 2026) - AA Math Index: 99.0 (source: Artificial Analysis, read October 22, 2026) - GPQA Diamond (Artificial Analysis run): 90.3% (source: Artificial Analysis, read October 22, 2026) - Humanity's Last Exam (Artificial Analysis run): 37.7% (source: Artificial Analysis, read October 22, 2026) - MMLU Pro (Artificial Analysis run): 87.4% (source: Artificial Analysis, read October 22, 2026) - LiveCodeBench (Artificial Analysis run): 89.4% (source: Artificial Analysis, read October 22, 2026) - Terminal-Bench Hard (Artificial Analysis run): 47.0% (source: Artificial Analysis, read October 22, 2026) - IFBench: 75.4% (source: Artificial Analysis, read October 22, 2026) - Long context reasoning (AA-LCR): 82.7% (source: Artificial Analysis, read October 22, 2026) - τ²-Bench Telecom: 84.8% (source: Artificial Analysis, read October 22, 2026) - AIME 2025 (Artificial Analysis run): 99.0% (source: Artificial Analysis, read October 22, 2026) - Terminal-Bench: 64.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - GSO-Bench: 27.5% (source: Epoch AI benchmarking hub, read October 22, 2026) - WeirdML: 72.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - METR time horizons: 75.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - Remote Labor Index: 2.5% (source: Epoch AI benchmarking hub, read October 22, 2026) - GDPval: 49.7% (source: Epoch AI benchmarking hub, read October 22, 2026) - DeepResearch Bench: 41.1% (source: Epoch AI benchmarking hub, read October 22, 2026) - GPQA Diamond (Epoch AI run): 91.4% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI: 86.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI-2: 52.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - SimpleBench: 45.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - Chess puzzles: 48.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - Mystery game puzzles: 23.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - EBR-bench: 23.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - OTIS Mock AIME: 96.1% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierMath, tiers 1 to 3: 67.4% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierMath, tier 4: 31.7% (source: Epoch AI benchmarking hub, read October 22, 2026) - ProofBench: 15.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - SimpleQA Verified: 37.1% (source: Epoch AI benchmarking hub, read October 22, 2026) - DTBench: 90.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - LMCA: 43.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - CL-bench: 18.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - Furniture assembly: 38.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - VPCT: 84.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - ALE-Bench: 1293.5 (source: ALE-Bench (Sakana AI), read October 22, 2026) - Vending-Bench 2: $3,591 (source: Epoch AI benchmarking hub, read October 22, 2026) - Epoch Capabilities Index: 153.7 (source: Epoch AI benchmarking hub, read October 22, 2026) - LMArena Text: 1416 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: coding: 1442 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts: 1428 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: math: 1440 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: instruction following: 1409 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: creative writing: 1376 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: 1246 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: English: 1422 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: non-English: 1405 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Chinese: 1447 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: German: 1444 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Japanese: 1420 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Korean: 1374 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: French: 1454 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Spanish: 1413 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Russian: 1415 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Polish: 1426 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: multi-turn: 1419 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: longer queries: 1413 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: expert: 1445 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts in English: 1431 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: software and IT: 1434 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: mathematical work: 1442 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: science: 1425 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: medicine and healthcare: 1425 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: legal and government: 1427 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: business and finance: 1414 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: writing and language: 1392 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: entertainment and media: 1378 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: HTML: 1427 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: English: 1243 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: Chinese: 1305 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: reading text: 1257 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: diagrams: 1265 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: homework: 1276 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: humor: 1227 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: captioning: 1267 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: recognizing things: 1232 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: creative writing: 1225 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Document: 1405 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - Berkeley Function Calling Leaderboard (BFCL): 55.9% (source: Berkeley Function Calling Leaderboard, read October 22, 2026) - Kagi LLM Benchmark: 73.3% (source: Kagi LLM Benchmarking Project, read October 22, 2026) - FACTS Benchmark Suite: 48.5% (source: Kaggle Benchmarks, read October 22, 2026)