# MoonshotAI Kimi K2.5 As of October 22, 2026, MoonshotAI Kimi K2.5 ranks #82 of 123 models overall with 44.8 ±11.7 on the LLMBenchmark Index. Kimi K2.5 ranks #51 of 133 in web design, its strongest category. MoonshotAI Kimi K2.5 costs $0.532 per million input tokens and $2.85 per million output tokens. Kimi K2.5 was released on January 27, 2026. Page: https://llmbenchmark.ai/models/kimi-k2-5 Data from October 22, 2026. Every figure below names the board it was read from and the day. - Maker: Moonshot AI - Released: January 27, 2026 - Price per million input tokens: $0.532 - Price per million output tokens: $2.85 - Context window: 262K tokens - Overall rank: 82 of 123, by the LLMBenchmark Index - LLMBenchmark Index: 44.8 ±11.7 (90% margin, from 5 of 7 inputs) - Artificial Analysis Intelligence Index: 23.5 - Vibe Coding Index: 38.6 (from 3 of 6 coding benchmarks) Benchmark results (99 of 195): - SWE-bench Verified: 70.8% (source: SWE-bench (swebench.com), read October 22, 2026) - SWE-bench Multilingual: 67.3% (source: SWE-bench (swebench.com), read October 22, 2026) - LMArena WebDev: 1437 Elo (source: LMArena WebDev, read October 22, 2026) - GPQA Diamond: 84.1% (source: GPQA (Vals AI), read October 22, 2026) - AIME: 95.6% (source: AIME (Vals AI), read October 22, 2026) - MMLU Pro: 85.9% (source: MMLU Pro (Vals AI), read October 22, 2026) - AA Intelligence Index: 23.5 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - AA Coding Index: 46.8 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - Design Arena: websites: 1248 Elo (source: Design Arena, read October 22, 2026) - Design Arena: UI components: 1234 Elo (source: Design Arena, read October 22, 2026) - Design Arena: game dev: 1206 Elo (source: Design Arena, read October 22, 2026) - Design Arena: data viz: 1219 Elo (source: Design Arena, read October 22, 2026) - Design Arena: 3D: 1204 Elo (source: Design Arena, read October 22, 2026) - Design Arena: all code categories: 1236 Elo (source: Design Arena, read October 22, 2026) - Design Arena: SVG: 1149 Elo (source: Design Arena, read October 22, 2026) - Design Arena: ASCII art: 1166 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building web apps: 1136 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building full-stack apps: 1114 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building mobile apps: 1147 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building Android apps: 1118 Elo (source: Design Arena, read October 22, 2026) - Design Arena: agents building Godot games: 1219 Elo (source: Design Arena, read October 22, 2026) - LiveCodeBench (Vals AI): 83.9% (source: Vals AI boards, read October 22, 2026) - SWE-bench Verified (Vals AI): 70.0% (source: Vals AI boards, read October 22, 2026) - Vibe Code Bench (Vals AI): 17.5% (source: Vals AI boards, read October 22, 2026) - Code migration (Vals AI): 7.0% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench 2.1 (Vals AI): 42.0% (source: Vals AI boards, read October 22, 2026) - Finance Agent (Vals AI): 35.8% (source: Vals AI boards, read October 22, 2026) - Legal Research Bench (Vals AI): 15.9% (source: Vals AI boards, read October 22, 2026) - Excel Modeling Benchmark (Vals AI): 28.5% (source: Vals AI boards, read October 22, 2026) - MMMU Pro (Vals AI): 84.3% (source: Vals AI boards, read October 22, 2026) - SAGE (Vals AI): 49.9% (source: Vals AI boards, read October 22, 2026) - CaseLaw (Vals AI): 58.7% (source: Vals AI boards, read October 22, 2026) - CorpFin (Vals AI): 68.3% (source: Vals AI boards, read October 22, 2026) - TaxEval (Vals AI): 74.2% (source: Vals AI boards, read October 22, 2026) - MortgageTax (Vals AI): 66.5% (source: Vals AI boards, read October 22, 2026) - MedQA (Vals AI): 94.4% (source: Vals AI boards, read October 22, 2026) - MedCode (Vals AI): 39.3% (source: Vals AI boards, read October 22, 2026) - MedScribe (Vals AI): 76.4% (source: Vals AI boards, read October 22, 2026) - GPQA Diamond (Artificial Analysis run): 87.9% (source: Artificial Analysis, read October 22, 2026) - Humanity's Last Exam (Artificial Analysis run): 30.7% (source: Artificial Analysis, read October 22, 2026) - Terminal-Bench Hard (Artificial Analysis run): 34.8% (source: Artificial Analysis, read October 22, 2026) - IFBench: 70.2% (source: Artificial Analysis, read October 22, 2026) - Long context reasoning (AA-LCR): 78.0% (source: Artificial Analysis, read October 22, 2026) - τ²-Bench Telecom: 95.9% (source: Artificial Analysis, read October 22, 2026) - Terminal-Bench: 43.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - WeirdML: 45.6% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI: 65.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI-2: 11.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - SimpleBench: 46.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - SimpleQA Verified: 34.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - Fiction.LiveBench: 86.1% (source: Epoch AI benchmarking hub, read October 22, 2026) - CL-bench: 19.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - OSWorld: 63.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - CL-bench Life: 13.2% (source: Epoch AI benchmarking hub, read October 22, 2026) - ALE-Bench: 821.6 (source: ALE-Bench (Sakana AI), read October 22, 2026) - Vending-Bench 2: $1,198 (source: Epoch AI benchmarking hub, read October 22, 2026) - Epoch Capabilities Index: 148.1 (source: Epoch AI benchmarking hub, read October 22, 2026) - LMArena Text: 1445 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: coding: 1483 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts: 1453 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: math: 1470 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: instruction following: 1431 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: creative writing: 1422 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: 1269 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: English: 1453 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: non-English: 1433 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Chinese: 1494 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: German: 1441 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Japanese: 1423 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Korean: 1412 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: 1450 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Russian: 1435 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Polish: 1445 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: multi-turn: 1444 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: longer queries: 1445 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: expert: 1466 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts in English: 1460 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: software and IT: 1471 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: mathematical work: 1473 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: science: 1459 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: medicine and healthcare: 1448 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: legal and government: 1441 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: business and finance: 1427 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: writing and language: 1425 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: entertainment and media: 1417 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: React: 1429 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: HTML: 1426 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: from an image: 1439 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: English: 1274 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: Chinese: 1328 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: reading text: 1276 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: diagrams: 1282 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: homework: 1297 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: humor: 1284 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: captioning: 1257 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: recognizing things: 1285 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Vision: creative writing: 1269 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - Kagi LLM Benchmark: 78.5% (source: Kagi LLM Benchmarking Project, read October 22, 2026)