# OpenAI GPT-5.6 Terra As of October 22, 2026, OpenAI GPT-5.6 Terra ranks #20 of 123 models overall with 68.1 ±7.2 on the LLMBenchmark Index. GPT-5.6 Terra ranks #10 of 108 in math, its strongest category. OpenAI GPT-5.6 Terra costs $2.00 per million input tokens and $12.00 per million output tokens. GPT-5.6 Terra was released on July 9, 2026. Page: https://llmbenchmark.ai/models/gpt-5-6-terra Data from October 22, 2026. Every figure below names the board it was read from and the day. - Maker: OpenAI - Released: July 9, 2026 - Price per million input tokens: $2.00 - Price per million output tokens: $12.00 - Context window: 1.05M tokens - Overall rank: 20 of 123, by the LLMBenchmark Index - LLMBenchmark Index: 68.1 ±7.2 (90% margin, from 7 of 7 inputs) - Artificial Analysis Intelligence Index: 42.1 - Vibe Coding Index: 63.9 (from 6 of 6 coding benchmarks) Benchmark results (127 of 195): - LiveBench coding: 86.2% (source: LiveBench, read October 22, 2026) - LMArena WebDev: 1522 Elo (source: LMArena WebDev, read October 22, 2026) - GPQA Diamond: 90.9% (source: GPQA (Vals AI), read October 22, 2026) - MMLU Pro: 86.7% (source: MMLU Pro (Vals AI), read October 22, 2026) - AA Intelligence Index: 42.1 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - AA Coding Index: 76.7 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - AA Agentic Index: 43.2 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - Design Arena: websites: 1268 Elo (source: Design Arena, read October 22, 2026) - Design Arena: UI components: 1256 Elo (source: Design Arena, read October 22, 2026) - Design Arena: game dev: 1229 Elo (source: Design Arena, read October 22, 2026) - Design Arena: data viz: 1240 Elo (source: Design Arena, read October 22, 2026) - Design Arena: 3D: 1134 Elo (source: Design Arena, read October 22, 2026) - Design Arena: all code categories: 1241 Elo (source: Design Arena, read October 22, 2026) - Design Arena: SVG: 1219 Elo (source: Design Arena, read October 22, 2026) - Design Arena: ASCII art: 1177 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building web apps: 1120 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building full-stack apps: 1061 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building mobile apps: 1111 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building Android apps: 1062 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building games: 1083 Elo (source: Design Arena, read October 22, 2026) - Design Arena: HTML slides: 1058 Elo (source: Design Arena, read October 22, 2026) - Design Arena: PowerPoint slides: 1105 Elo (source: Design Arena, read October 22, 2026) - LiveBench reasoning: 90.6% (source: LiveBench, read October 22, 2026) - LiveBench data analysis: 79.3% (source: LiveBench, read October 22, 2026) - LiveBench research: 60.4% (source: LiveBench, read October 22, 2026) - LiveBench browser: 84.0% (source: LiveBench, read October 22, 2026) - LiveBench long agentic: 48.4% (source: LiveBench, read October 22, 2026) - LiveBench design: 76.0% (source: LiveBench, read October 22, 2026) - LiveCodeBench (Vals AI): 85.9% (source: Vals AI boards, read October 22, 2026) - SWE-bench Verified (Vals AI): 95.4% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench 4 (Vals AI): 22.7% (source: Vals AI boards, read October 22, 2026) - Vibe Code Bench (Vals AI): 74.6% (source: Vals AI boards, read October 22, 2026) - Code migration (Vals AI): 47.8% (source: Vals AI boards, read October 22, 2026) - ProgramBench (Vals AI): 0.5% (source: Vals AI boards, read October 22, 2026) - SREBench (Vals AI): 9.5% (source: Vals AI boards, read October 22, 2026) - IOI (Vals AI): 87.6% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench 2.1 (Vals AI): 77.5% (source: Vals AI boards, read October 22, 2026) - Vibe Code Bench 1-100 (Vals AI): 14.8% (source: Vals AI boards, read October 22, 2026) - SkillsBench (Vals AI): 58.9% (source: Vals AI boards, read October 22, 2026) - CyberBench (Vals AI): 72.1% (source: Vals AI boards, read October 22, 2026) - Finance Agent (Vals AI): 54.4% (source: Vals AI boards, read October 22, 2026) - Tax Agent Bench (Vals AI): 30.0% (source: Vals AI boards, read October 22, 2026) - Legal Research Bench (Vals AI): 41.4% (source: Vals AI boards, read October 22, 2026) - Excel Modeling Benchmark (Vals AI): 66.2% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench-Science (Vals AI): 10.0% (source: Vals AI boards, read October 22, 2026) - Public Benefits Bench (Vals AI): 62.4% (source: Vals AI boards, read October 22, 2026) - MMMU Pro (Vals AI): 86.5% (source: Vals AI boards, read October 22, 2026) - SAGE (Vals AI): 47.0% (source: Vals AI boards, read October 22, 2026) - LegalBench (Vals AI): 85.1% (source: Vals AI boards, read October 22, 2026) - CorpFin (Vals AI): 65.3% (source: Vals AI boards, read October 22, 2026) - TaxEval (Vals AI): 76.2% (source: Vals AI boards, read October 22, 2026) - MortgageTax (Vals AI): 67.3% (source: Vals AI boards, read October 22, 2026) - MedCode (Vals AI): 43.4% (source: Vals AI boards, read October 22, 2026) - MedScribe (Vals AI): 82.9% (source: Vals AI boards, read October 22, 2026) - Vals Multimodal Index: 65.1% (source: Vals AI boards, read October 22, 2026) - GPQA Diamond (Artificial Analysis run): 92.5% (source: Artificial Analysis, read October 22, 2026) - Humanity's Last Exam (Artificial Analysis run): 42.9% (source: Artificial Analysis, read October 22, 2026) - SciCode: 55.0% (source: Artificial Analysis, read October 22, 2026) - Terminal-Bench Hard (Artificial Analysis run): 62.9% (source: Artificial Analysis, read October 22, 2026) - IFBench: 71.2% (source: Artificial Analysis, read October 22, 2026) - Long context reasoning (AA-LCR): 83.0% (source: Artificial Analysis, read October 22, 2026) - τ²-Bench Telecom: 86.3% (source: Artificial Analysis, read October 22, 2026) - FrontierCode: 41.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - WeirdML: 78.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - APEX-Agents: 58.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - BALROG: 53.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - GPQA Diamond (Epoch AI run): 93.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI: 96.5% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI-2: 83.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - SimpleBench: 48.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - Chess puzzles: 47.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - Mystery game puzzles: 35.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - OTIS Mock AIME: 99.7% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierMath, tiers 1 to 3: 86.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierMath, tier 4: 70.7% (source: Epoch AI benchmarking hub, read October 22, 2026) - ProofBench: 74.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - SimpleQA Verified: 43.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - DTBench: 93.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - LMCA: 54.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - Surface Evolver Bench: 83.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - Furniture assembly: 54.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - ALE-Bench: 1951.4 (source: ALE-Bench (Sakana AI), read October 22, 2026) - Vending-Bench 2: $7,343 (source: Epoch AI benchmarking hub, read October 22, 2026) - Epoch Capabilities Index: 160.1 (source: Epoch AI benchmarking hub, read October 22, 2026) - LMArena Text: 1448 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: coding: 1485 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts: 1469 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: math: 1465 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: instruction following: 1456 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: creative writing: 1413 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: 1271 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: English: 1449 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: non-English: 1441 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Chinese: 1514 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: German: 1460 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Japanese: 1454 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Korean: 1427 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: French: 1466 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Spanish: 1451 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Russian: 1452 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Polish: 1446 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: multi-turn: 1453 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: longer queries: 1454 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: expert: 1496 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts in English: 1465 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: software and IT: 1482 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: mathematical work: 1478 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: science: 1452 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: medicine and healthcare: 1430 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: legal and government: 1461 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: business and finance: 1448 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: writing and language: 1431 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: entertainment and media: 1418 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: React: 1513 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: HTML: 1526 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: from an image: 1541 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: English: 1272 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: Chinese: 1337 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: reading text: 1278 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: diagrams: 1295 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: homework: 1312 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: humor: 1236 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: creative writing: 1248 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Document: 1472 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - Kagi LLM Benchmark: 51.3% (source: Kagi LLM Benchmarking Project, read October 22, 2026) - SciCode: 10.8% (source: Kaggle Benchmarks, read October 22, 2026) - FACTS Benchmark Suite: 52.0% (source: Kaggle Benchmarks, read October 22, 2026)