# MoonshotAI Kimi K3 As of October 21, 2026, MoonshotAI Kimi K3 ranks #18 of 213 models overall with 43.6 on the Artificial Analysis Intelligence Index. Its strongest category is web design, where it ranks #4 of 133. MoonshotAI Kimi K3 costs $0.239 per million input tokens and $11.936 per million output tokens. It was released on July 16, 2026. Page: https://llmbenchmark.ai/models/kimi-k3 Data from October 21, 2026. Every figure below names the board it was read from and the day. - Maker: Moonshot AI - Released: July 16, 2026 - Price per million input tokens: $0.239 - Price per million output tokens: $11.936 - Context window: 1.05M tokens - Overall rank: 18 of 213, by the Artificial Analysis Intelligence Index - Artificial Analysis Intelligence Index: 43.6 - Vibe Coding Index: 61.8 (from 2 of 4 coding benchmarks) Benchmark results (117 of 190): - LiveBench coding: 59.3% (source: LiveBench, read October 21, 2026) - LMArena WebDev: 1655 Elo (source: LMArena WebDev, read October 21, 2026) - Vals Index: 53.0% (source: Vals Index, read October 21, 2026) - GPQA Diamond: 92.9% (source: GPQA (Vals AI), read October 21, 2026) - MMLU Pro: 88.0% (source: MMLU Pro (Vals AI), read October 21, 2026) - AA Intelligence Index: 43.6 (source: Artificial Analysis (via OpenRouter), read October 21, 2026) - AA Coding Index: 76.2 (source: Artificial Analysis (via OpenRouter), read October 21, 2026) - AA Agentic Index: 50.0 (source: Artificial Analysis (via OpenRouter), read October 21, 2026) - Design Arena: websites: 1336 Elo (source: Design Arena, read October 21, 2026) - Design Arena: UI components: 1345 Elo (source: Design Arena, read October 21, 2026) - Design Arena: game dev: 1361 Elo (source: Design Arena, read October 21, 2026) - Design Arena: data viz: 1350 Elo (source: Design Arena, read October 21, 2026) - Design Arena: 3D: 1388 Elo (source: Design Arena, read October 21, 2026) - Design Arena: all code categories: 1361 Elo (source: Design Arena, read October 21, 2026) - Design Arena: SVG: 1301 Elo (source: Design Arena, read October 21, 2026) - Design Arena: agents building web apps: 1296 Elo (source: Design Arena, read October 21, 2026) - Design Arena: agents building full-stack apps: 1315 Elo (source: Design Arena, read October 21, 2026) - Design Arena: agents building mobile apps: 1267 Elo (source: Design Arena, read October 21, 2026) - Design Arena: agents building Android apps: 1256 Elo (source: Design Arena, read October 21, 2026) - Design Arena: agents building Godot games: 1223 Elo (source: Design Arena, read October 21, 2026) - Design Arena: agents building games: 1252 Elo (source: Design Arena, read October 21, 2026) - Design Arena: HTML slides: 1256 Elo (source: Design Arena, read October 21, 2026) - Design Arena: PowerPoint slides: 1273 Elo (source: Design Arena, read October 21, 2026) - Design Arena: spreadsheets: 1058 Elo (source: Design Arena, read October 21, 2026) - LiveBench reasoning: 90.7% (source: LiveBench, read October 21, 2026) - LiveBench data analysis: 78.7% (source: LiveBench, read October 21, 2026) - LiveBench research: 79.0% (source: LiveBench, read October 21, 2026) - LiveBench browser: 16.0% (source: LiveBench, read October 21, 2026) - LiveBench long agentic: 42.8% (source: LiveBench, read October 21, 2026) - LiveBench design: 77.2% (source: LiveBench, read October 21, 2026) - LiveCodeBench (Vals AI): 87.2% (source: Vals AI boards, read October 21, 2026) - SWE-bench Verified (Vals AI): 93.4% (source: Vals AI boards, read October 21, 2026) - Terminal-Bench 4 (Vals AI): 17.2% (source: Vals AI boards, read October 21, 2026) - Vibe Code Bench (Vals AI): 85.0% (source: Vals AI boards, read October 21, 2026) - Code migration (Vals AI): 16.1% (source: Vals AI boards, read October 21, 2026) - ProgramBench (Vals AI): 2.0% (source: Vals AI boards, read October 21, 2026) - IOI (Vals AI): 48.9% (source: Vals AI boards, read October 21, 2026) - Terminal-Bench 2.1 (Vals AI): 80.9% (source: Vals AI boards, read October 21, 2026) - SWE-bench Multimodal (Vals AI): 44.2% (source: Vals AI boards, read October 21, 2026) - Vibe Code Bench 1-100 (Vals AI): 18.2% (source: Vals AI boards, read October 21, 2026) - CyberBench (Vals AI): 75.2% (source: Vals AI boards, read October 21, 2026) - Finance Agent (Vals AI): 53.1% (source: Vals AI boards, read October 21, 2026) - Tax Agent Bench (Vals AI): 33.4% (source: Vals AI boards, read October 21, 2026) - Legal Research Bench (Vals AI): 46.1% (source: Vals AI boards, read October 21, 2026) - Harvey's Legal Agent Benchmark (Vals AI): 12.9% (source: Vals AI boards, read October 21, 2026) - Excel Modeling Benchmark (Vals AI): 66.7% (source: Vals AI boards, read October 21, 2026) - BioMysteryBench (Vals AI): 72.6% (source: Vals AI boards, read October 21, 2026) - Terminal-Bench-Science (Vals AI): 1.4% (source: Vals AI boards, read October 21, 2026) - Time Horizon Index (Vals AI): 10.5% (source: Vals AI boards, read October 21, 2026) - Public Benefits Bench (Vals AI): 68.2% (source: Vals AI boards, read October 21, 2026) - MMMU Pro (Vals AI): 88.2% (source: Vals AI boards, read October 21, 2026) - SAGE (Vals AI): 52.8% (source: Vals AI boards, read October 21, 2026) - LegalBench (Vals AI): 86.2% (source: Vals AI boards, read October 21, 2026) - CorpFin (Vals AI): 71.6% (source: Vals AI boards, read October 21, 2026) - TaxEval (Vals AI): 75.7% (source: Vals AI boards, read October 21, 2026) - MortgageTax (Vals AI): 66.3% (source: Vals AI boards, read October 21, 2026) - MedCode (Vals AI): 49.4% (source: Vals AI boards, read October 21, 2026) - MedScribe (Vals AI): 88.0% (source: Vals AI boards, read October 21, 2026) - Vals Multimodal Index: 73.4% (source: Vals AI boards, read October 21, 2026) - GPQA Diamond (Artificial Analysis run): 93.5% (source: Artificial Analysis, read October 21, 2026) - Humanity's Last Exam (Artificial Analysis run): 46.9% (source: Artificial Analysis, read October 21, 2026) - SciCode: 59.5% (source: Artificial Analysis, read October 21, 2026) - Long context reasoning (AA-LCR): 88.7% (source: Artificial Analysis, read October 21, 2026) - FrontierCode: 44.2% (source: Epoch AI benchmarking hub, read October 21, 2026) - FrontierSWE: 25.9% (source: Epoch AI benchmarking hub, read October 21, 2026) - WeirdML: 82.6% (source: Epoch AI benchmarking hub, read October 21, 2026) - APEX-Agents: 47.8% (source: Epoch AI benchmarking hub, read October 21, 2026) - PostTrainBench: 32.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - GPQA Diamond (Epoch AI run): 93.1% (source: Epoch AI benchmarking hub, read October 21, 2026) - ARC-AGI: 94.5% (source: Epoch AI benchmarking hub, read October 21, 2026) - ARC-AGI-2: 60.4% (source: Epoch AI benchmarking hub, read October 21, 2026) - SimpleBench: 60.7% (source: Epoch AI benchmarking hub, read October 21, 2026) - Chess puzzles: 37.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - Mystery game puzzles: 26.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - OTIS Mock AIME: 97.2% (source: Epoch AI benchmarking hub, read October 21, 2026) - FrontierMath, tiers 1 to 3: 72.2% (source: Epoch AI benchmarking hub, read October 21, 2026) - FrontierMath, tier 4: 39.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - ProofBench: 87.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - SimpleQA Verified: 50.6% (source: Epoch AI benchmarking hub, read October 21, 2026) - DTBench: 91.2% (source: Epoch AI benchmarking hub, read October 21, 2026) - LMCA: 52.7% (source: Epoch AI benchmarking hub, read October 21, 2026) - Surface Evolver Bench: 95.0% (source: Epoch AI benchmarking hub, read October 21, 2026) - Furniture assembly: 34.2% (source: Epoch AI benchmarking hub, read October 21, 2026) - ALE-Bench: 1524.5 (source: ALE-Bench (Sakana AI), read October 21, 2026) - Vending-Bench 2: $5,165 (source: Epoch AI benchmarking hub, read October 21, 2026) - Epoch Capabilities Index: 157.9 (source: Epoch AI benchmarking hub, read October 21, 2026) - LMArena Text: 1476 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: coding: 1508 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: hard prompts: 1495 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: math: 1494 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: instruction following: 1485 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: creative writing: 1457 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: English: 1481 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: non-English: 1466 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Chinese: 1529 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: German: 1490 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Japanese: 1494 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Korean: 1460 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: French: 1492 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Spanish: 1472 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Russian: 1485 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: Polish: 1458 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: multi-turn: 1489 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: longer queries: 1494 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: expert: 1520 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: hard prompts in English: 1496 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: software and IT: 1499 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: mathematical work: 1505 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: science: 1508 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: medicine and healthcare: 1489 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: legal and government: 1503 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: business and finance: 1470 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: writing and language: 1471 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena Text: entertainment and media: 1453 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena WebDev: React: 1670 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena WebDev: HTML: 1644 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026) - LMArena WebDev: from an image: 1579 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)