# Google Gemini 3.1 Pro Preview As of October 22, 2026, Google Gemini 3.1 Pro Preview ranks #60 of 123 models overall with 54.0 ±17.4 on the LLMBenchmark Index. Gemini 3.1 Pro Preview ranks #6 of 75 in knowledge, its strongest category. Google Gemini 3.1 Pro Preview costs $2.00 per million input tokens and $12.00 per million output tokens. Gemini 3.1 Pro Preview was released on February 19, 2026. Page: https://llmbenchmark.ai/models/gemini-3-1-pro-preview Data from October 22, 2026. Every figure below names the board it was read from and the day. - Maker: Google - Released: February 19, 2026 - Price per million input tokens: $2.00 - Price per million output tokens: $12.00 - Context window: 1.05M tokens - Overall rank: 60 of 123, by the LLMBenchmark Index - LLMBenchmark Index: 54.0 ±17.4 (90% margin, from 7 of 7 inputs) - Artificial Analysis Intelligence Index: 29.7 - Vibe Coding Index: 32.3 (from 4 of 6 coding benchmarks) Benchmark results (133 of 195): - LMArena WebDev: 1447 Elo (source: LMArena WebDev, read October 22, 2026) - GPQA Diamond: 95.5% (source: GPQA (Vals AI), read October 22, 2026) - AIME: 98.1% (source: AIME (Vals AI), read October 22, 2026) - MMLU Pro: 91.0% (source: MMLU Pro (Vals AI), read October 22, 2026) - AA Intelligence Index: 29.7 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - AA Coding Index: 68.8 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - AA Agentic Index: 8.2 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - Design Arena: websites: 1233 Elo (source: Design Arena, read October 22, 2026) - Design Arena: UI components: 1224 Elo (source: Design Arena, read October 22, 2026) - Design Arena: game dev: 1202 Elo (source: Design Arena, read October 22, 2026) - Design Arena: data viz: 1209 Elo (source: Design Arena, read October 22, 2026) - Design Arena: 3D: 1234 Elo (source: Design Arena, read October 22, 2026) - Design Arena: all code categories: 1231 Elo (source: Design Arena, read October 22, 2026) - Design Arena: SVG: 1285 Elo (source: Design Arena, read October 22, 2026) - Design Arena: ASCII art: 1270 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building web apps: 1125 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building full-stack apps: 1068 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building mobile apps: 1094 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building Android apps: 1049 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building Godot games: 1236 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building games: 1113 Elo (source: Design Arena, read October 22, 2026) - Design Arena: HTML slides: 1103 Elo (source: Design Arena, read October 22, 2026) - Design Arena: PowerPoint slides: 1041 Elo (source: Design Arena, read October 22, 2026) - LiveCodeBench (Vals AI): 88.5% (source: Vals AI boards, read October 22, 2026) - SWE-bench Verified (Vals AI): 78.8% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench 4 (Vals AI): 2.5% (source: Vals AI boards, read October 22, 2026) - Vibe Code Bench (Vals AI): 32.0% (source: Vals AI boards, read October 22, 2026) - Code migration (Vals AI): 17.3% (source: Vals AI boards, read October 22, 2026) - ProgramBench (Vals AI): 0.0% (source: Vals AI boards, read October 22, 2026) - SREBench (Vals AI): 0.0% (source: Vals AI boards, read October 22, 2026) - IOI (Vals AI): 51.8% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench 2.1 (Vals AI): 70.8% (source: Vals AI boards, read October 22, 2026) - Vibe Code Bench 1-100 (Vals AI): 6.7% (source: Vals AI boards, read October 22, 2026) - Finance Agent (Vals AI): 43.0% (source: Vals AI boards, read October 22, 2026) - Tax Agent Bench (Vals AI): 9.3% (source: Vals AI boards, read October 22, 2026) - Legal Research Bench (Vals AI): 20.7% (source: Vals AI boards, read October 22, 2026) - Excel Modeling Benchmark (Vals AI): 52.6% (source: Vals AI boards, read October 22, 2026) - BioMysteryBench (Vals AI): 60.0% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench-Science (Vals AI): 1.4% (source: Vals AI boards, read October 22, 2026) - Vals Science Index: 27.4% (source: Vals AI boards, read October 22, 2026) - Public Benefits Bench (Vals AI): 53.8% (source: Vals AI boards, read October 22, 2026) - MysteryMechanism (Vals AI): 20.7% (source: Vals AI boards, read October 22, 2026) - MMMU Pro (Vals AI): 88.2% (source: Vals AI boards, read October 22, 2026) - SAGE (Vals AI): 48.7% (source: Vals AI boards, read October 22, 2026) - LegalBench (Vals AI): 87.4% (source: Vals AI boards, read October 22, 2026) - CaseLaw (Vals AI): 64.8% (source: Vals AI boards, read October 22, 2026) - CorpFin (Vals AI): 64.5% (source: Vals AI boards, read October 22, 2026) - TaxEval (Vals AI): 72.9% (source: Vals AI boards, read October 22, 2026) - MortgageTax (Vals AI): 69.4% (source: Vals AI boards, read October 22, 2026) - MedQA (Vals AI): 96.4% (source: Vals AI boards, read October 22, 2026) - MedCode (Vals AI): 59.1% (source: Vals AI boards, read October 22, 2026) - MedScribe (Vals AI): 76.1% (source: Vals AI boards, read October 22, 2026) - Vals Multimodal Index: 56.1% (source: Vals AI boards, read October 22, 2026) - GPQA Diamond (Artificial Analysis run): 94.1% (source: Artificial Analysis, read October 22, 2026) - Humanity's Last Exam (Artificial Analysis run): 47.0% (source: Artificial Analysis, read October 22, 2026) - SciCode: 58.7% (source: Artificial Analysis, read October 22, 2026) - Terminal-Bench Hard (Artificial Analysis run): 53.8% (source: Artificial Analysis, read October 22, 2026) - IFBench: 77.1% (source: Artificial Analysis, read October 22, 2026) - Long context reasoning (AA-LCR): 82.0% (source: Artificial Analysis, read October 22, 2026) - τ²-Bench Telecom: 95.6% (source: Artificial Analysis, read October 22, 2026) - Terminal-Bench: 80.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - GSO-Bench: 22.6% (source: Epoch AI benchmarking hub, read October 22, 2026) - WeirdML: 72.1% (source: Epoch AI benchmarking hub, read October 22, 2026) - MirrorCode: 8.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - APEX-Agents: 35.1% (source: Epoch AI benchmarking hub, read October 22, 2026) - METR time horizons: 77.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - BALROG: 57.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - DeepResearch Bench: 47.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - PostTrainBench: 22.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - GPQA Diamond (Epoch AI run): 94.4% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI: 98.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI-2: 77.1% (source: Epoch AI benchmarking hub, read October 22, 2026) - SimpleBench: 79.6% (source: Epoch AI benchmarking hub, read October 22, 2026) - Chess puzzles: 51.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - Mystery game puzzles: 34.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - EBR-bench: 14.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - OTIS Mock AIME: 95.6% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierMath, tiers 1 to 3: 59.6% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierMath, tier 4: 26.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - ProofBench: 26.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - SimpleQA Verified: 73.5% (source: Epoch AI benchmarking hub, read October 22, 2026) - DTBench: 97.1% (source: Epoch AI benchmarking hub, read October 22, 2026) - LMCA: 53.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - CL-bench: 20.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - Furniture assembly: 26.7% (source: Epoch AI benchmarking hub, read October 22, 2026) - ExploitBench: 26.1% (source: Epoch AI benchmarking hub, read October 22, 2026) - CL-bench Life: 16.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - ALE-Bench: 1160.6 (source: ALE-Bench (Sakana AI), read October 22, 2026) - Vending-Bench 2: $911 (source: Epoch AI benchmarking hub, read October 22, 2026) - LMArena Text: 1480 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: coding: 1484 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts: 1485 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: math: 1486 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: instruction following: 1466 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: creative writing: 1482 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: 1296 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: English: 1478 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: non-English: 1476 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Chinese: 1527 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: German: 1490 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Japanese: 1491 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Korean: 1455 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: French: 1487 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Spanish: 1481 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Russian: 1497 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Polish: 1498 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: multi-turn: 1487 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: longer queries: 1482 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: expert: 1483 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts in English: 1480 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: software and IT: 1487 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: 1497 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: medicine and healthcare: 1486 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: legal and government: 1487 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: business and finance: 1459 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: writing and language: 1481 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: entertainment and media: 1465 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: React: 1438 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: HTML: 1490 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: from an image: 1481 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: English: 1287 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: Chinese: 1364 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: reading text: 1304 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: diagrams: 1311 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: homework: 1316 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: humor: 1322 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: captioning: 1345 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: recognizing things: 1330 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: creative writing: 1324 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Document: 1444 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - SciCode: 14.3% (source: Kaggle Benchmarks, read October 22, 2026) - FACTS Benchmark Suite: 67.4% (source: Kaggle Benchmarks, read October 22, 2026)