# Google Gemini 3.8 Flash As of October 22, 2026, Google Gemini 3.8 Flash ranks #15 of 123 models overall with 69.0 ±9.6 on the LLMBenchmark Index. Gemini 3.8 Flash ranks #11 of 122 in reasoning, its strongest category. Google Gemini 3.8 Flash costs $0.75 per million input tokens and $3.75 per million output tokens. Gemini 3.8 Flash was released on September 2, 2026. Page: https://llmbenchmark.ai/models/gemini-3-8-flash Data from October 22, 2026. Every figure below names the board it was read from and the day. - Maker: Google - Released: September 2, 2026 - Price per million input tokens: $0.75 - Price per million output tokens: $3.75 - Context window: 1.05M tokens - Overall rank: 15 of 123, by the LLMBenchmark Index - LLMBenchmark Index: 69.0 ±9.6 (90% margin, from 7 of 7 inputs) - Artificial Analysis Intelligence Index: 40.9 - Vibe Coding Index: 61.6 (from 6 of 6 coding benchmarks) Benchmark results (128 of 195): - LiveBench coding: 82.2% (source: LiveBench, read October 22, 2026) - LMArena WebDev: 1585 Elo (source: LMArena WebDev, read October 22, 2026) - Vals Index: 56.0% (source: Vals Index, read October 22, 2026) - GPQA Diamond: 94.4% (source: GPQA (Vals AI), read October 22, 2026) - MMLU Pro: 90.2% (source: MMLU Pro (Vals AI), read October 22, 2026) - AA Intelligence Index: 40.9 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - AA Coding Index: 76.3 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - AA Agentic Index: 40.2 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - Design Arena: websites: 1300 Elo (source: Design Arena, read October 22, 2026) - Design Arena: UI components: 1316 Elo (source: Design Arena, read October 22, 2026) - Design Arena: game dev: 1304 Elo (source: Design Arena, read October 22, 2026) - Design Arena: data viz: 1270 Elo (source: Design Arena, read October 22, 2026) - Design Arena: 3D: 1294 Elo (source: Design Arena, read October 22, 2026) - Design Arena: all code categories: 1303 Elo (source: Design Arena, read October 22, 2026) - Design Arena: SVG: 1296 Elo (source: Design Arena, read October 22, 2026) - Design Arena: ASCII art: 1262 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building web apps: 1238 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building full-stack apps: 1244 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building mobile apps: 1236 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building Android apps: 1078 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building Godot games: 1273 Elo (source: Design Arena, read October 22, 2026) - Design Arena: PowerPoint slides: 1153 Elo (source: Design Arena, read October 22, 2026) - Design Arena: spreadsheets: 1022 Elo (source: Design Arena, read October 22, 2026) - LiveBench reasoning: 89.3% (source: LiveBench, read October 22, 2026) - LiveBench data analysis: 54.0% (source: LiveBench, read October 22, 2026) - LiveBench research: 79.1% (source: LiveBench, read October 22, 2026) - LiveBench browser: 64.0% (source: LiveBench, read October 22, 2026) - LiveBench long agentic: 30.1% (source: LiveBench, read October 22, 2026) - LiveBench design: 72.7% (source: LiveBench, read October 22, 2026) - LiveCodeBench (Vals AI): 89.5% (source: Vals AI boards, read October 22, 2026) - SWE-bench Verified (Vals AI): 80.0% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench 4 (Vals AI): 19.7% (source: Vals AI boards, read October 22, 2026) - Vibe Code Bench (Vals AI): 78.7% (source: Vals AI boards, read October 22, 2026) - Code migration (Vals AI): 36.5% (source: Vals AI boards, read October 22, 2026) - ProgramBench (Vals AI): 1.0% (source: Vals AI boards, read October 22, 2026) - SREBench (Vals AI): 11.1% (source: Vals AI boards, read October 22, 2026) - IOI (Vals AI): 56.9% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench 2.1 (Vals AI): 81.3% (source: Vals AI boards, read October 22, 2026) - SWE-bench Multimodal (Vals AI): 40.8% (source: Vals AI boards, read October 22, 2026) - Vibe Code Bench 1-100 (Vals AI): 18.8% (source: Vals AI boards, read October 22, 2026) - SkillsBench (Vals AI): 58.0% (source: Vals AI boards, read October 22, 2026) - CyberBench (Vals AI): 43.8% (source: Vals AI boards, read October 22, 2026) - Finance Agent (Vals AI): 61.4% (source: Vals AI boards, read October 22, 2026) - Tax Agent Bench (Vals AI): 32.1% (source: Vals AI boards, read October 22, 2026) - Legal Research Bench (Vals AI): 38.9% (source: Vals AI boards, read October 22, 2026) - Harvey's Legal Agent Benchmark (Vals AI): 12.1% (source: Vals AI boards, read October 22, 2026) - Excel Modeling Benchmark (Vals AI): 72.2% (source: Vals AI boards, read October 22, 2026) - BioMysteryBench (Vals AI): 62.2% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench-Science (Vals AI): 8.6% (source: Vals AI boards, read October 22, 2026) - Vals Science Index: 35.8% (source: Vals AI boards, read October 22, 2026) - Time Horizon Index (Vals AI): 18.8% (source: Vals AI boards, read October 22, 2026) - CUA-bench (Vals AI): 4.2% (source: Vals AI boards, read October 22, 2026) - Public Benefits Bench (Vals AI): 65.3% (source: Vals AI boards, read October 22, 2026) - MysteryMechanism (Vals AI): 36.5% (source: Vals AI boards, read October 22, 2026) - MMMU Pro (Vals AI): 89.1% (source: Vals AI boards, read October 22, 2026) - SAGE (Vals AI): 35.1% (source: Vals AI boards, read October 22, 2026) - LegalBench (Vals AI): 87.0% (source: Vals AI boards, read October 22, 2026) - TaxEval (Vals AI): 74.5% (source: Vals AI boards, read October 22, 2026) - MortgageTax (Vals AI): 65.3% (source: Vals AI boards, read October 22, 2026) - MedCode (Vals AI): 48.1% (source: Vals AI boards, read October 22, 2026) - MedScribe (Vals AI): 84.5% (source: Vals AI boards, read October 22, 2026) - GPQA Diamond (Artificial Analysis run): 95.3% (source: Artificial Analysis, read October 22, 2026) - Humanity's Last Exam (Artificial Analysis run): 47.8% (source: Artificial Analysis, read October 22, 2026) - SciCode: 56.6% (source: Artificial Analysis, read October 22, 2026) - Long context reasoning (AA-LCR): 84.0% (source: Artificial Analysis, read October 22, 2026) - FrontierCode: 41.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierSWE: 19.6% (source: Epoch AI benchmarking hub, read October 22, 2026) - WeirdML: 84.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - APEX-Agents: 64.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - Remote Labor Index: 5.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - GPQA Diamond (Epoch AI run): 95.4% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI: 98.5% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI-2: 89.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - Chess puzzles: 58.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - Mystery game puzzles: 47.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - OTIS Mock AIME: 98.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierMath, tiers 1 to 3: 68.4% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierMath, tier 4: 22.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - ProofBench: 48.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - SimpleQA Verified: 69.7% (source: Epoch AI benchmarking hub, read October 22, 2026) - DTBench: 95.7% (source: Epoch AI benchmarking hub, read October 22, 2026) - LMCA: 52.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - Surface Evolver Bench: 76.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - Furniture assembly: 31.7% (source: Epoch AI benchmarking hub, read October 22, 2026) - ALE-Bench: 1269.7 (source: ALE-Bench (Sakana AI), read October 22, 2026) - Vending-Bench 2: $5,094 (source: Epoch AI benchmarking hub, read October 22, 2026) - Epoch Capabilities Index: 157.4 (source: Epoch AI benchmarking hub, read October 22, 2026) - LMArena Text: 1499 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: coding: 1508 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts: 1509 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: math: 1532 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: instruction following: 1492 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: creative writing: 1488 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: 1315 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: English: 1498 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: non-English: 1491 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Chinese: 1553 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: German: 1501 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Japanese: 1507 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Korean: 1464 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: French: 1496 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Spanish: 1487 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Russian: 1514 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Polish: 1497 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: multi-turn: 1503 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: longer queries: 1507 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: expert: 1526 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts in English: 1504 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: software and IT: 1507 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: mathematical work: 1519 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: science: 1523 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: medicine and healthcare: 1508 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: legal and government: 1497 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: business and finance: 1479 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: writing and language: 1494 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: entertainment and media: 1480 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: React: 1577 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: HTML: 1616 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: from an image: 1593 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: English: 1318 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: Chinese: 1416 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: reading text: 1333 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: diagrams: 1345 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: homework: 1351 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: humor: 1321 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: creative writing: 1299 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - SciCode: 12.3% (source: Kaggle Benchmarks, read October 22, 2026) - FACTS Benchmark Suite: 69.4% (source: Kaggle Benchmarks, read October 22, 2026)