Vascend
HorizonVascend IDEVascend CLICommunityBenchmarkContacto
AccederÚnete a Discord
InicioHorizonIDECLIAcceder
ChangelogInteligenciaCoste por tareaCoding Agent IndexCyber IndexDecision IndexBenchmarksVelocidadLatenciaPrecioClasificación
Benchmark

Kimi K3.1 Pro

  • Kimi
  • Lanzamiento · 23 de octubre de 2026
  • Propietario
  • Razonamiento

Inteligencia

Intelligence Index

#5 de 266
  1. 57,658Claude Opus 5.5 (Max)
  2. 5656Claude Opus 5.5 (Xhigh)
  3. 5656Claude Sonnet 5.5 (Max)
  4. 55,856Muse Wonder Full (Max)
  5. 55,155Kimi K3.1 Pro
  6. 53,654Claude Opus 5.5 (High)
  7. 53,453Claude Fable 5.1 (Max)
  8. 53,253Claude Fable 5.1 (Xhigh)
  9. 52,753GPT-6 Astra (Max)
  10. 52,653Gemini 4 Argon (High)
Datos de Artificial Analysis

Evaluaciones

  • AA-Briefcase1678#4
  • AA-Omniscience Accuracy51%#30
  • AA-Omniscience Non-Hallucination56,1%#16
  • AutomationBench-AA73,3%#2
  • GDPval-AA1846#3
  • Humanity's Last Exam52,5%#18
  • GDP.pdf31,4%#2
  • Terminal-Bench 4.067,2%#1
  • AA-LCR86,3%#2
  • SciCode60,8%#9
  • CritPt24,6%#29
  • Terminal-Bench-Science34,3%#8
  • FrontierCode47,1%#4
  • AA-AnalystAgent38,8%#14
  • AA-3D1110#7
  • MMMU-Pro83,9%#17
  • PersonalityBench: Expressiveness62,9#14
  • PersonalityBench: Organization87,6#5
  • PersonalityBench: Transparency87,6#6
  • PersonalityBench: Slop20,8#19
  • PersonalityBench: Informativeness23,3#22
  • PersonalityBench: Warmth50,5#15
  • PersonalityBench: Conviction62,9#11
  • PersonalityBench: Imagination53#15
Kimi K3.1 ProMediana
Datos de Artificial Analysis

Coste por tarea

Coste por tarea

$ por tarea del Intelligence Index · #17 de 24
  1. $1,031,03Step 5 Preview
  2. $1,131,13Mistral Large 4 Preview
  3. $1,241,24Gemini 3.8 Flash (High)
  4. $1,991,99Gemini 4 Argon (High)
  5. $2,012,01GLM-5.3 (Max)
  6. $2,112,11Kimi K3.1 Pro
  7. $3,263,26GPT-6 Astra (Max)
  8. $3,743,74Grok 4.7 (Xhigh)
  9. $4,544,54Muse Wonder (Max)
  10. $5,415,41Qwen3.8 Max (0902)
Datos de Artificial Analysis

Tiempo por tarea

segundos por tarea · #21 de 26
  1. 628628GPT-6.1 Sol (Max)
  2. 723723Gemini 4 Argon (High)
  3. 736736Claude Fable 5.1 (Max)
  4. 791791Claude Opus 5.5 (Max)
  5. 824824GLM-5.3-Flash
  6. 846846Kimi K3.1 Pro
  7. 901901Grok 4.7 (Xhigh)
  8. 936936Claude Sonnet 5.5 (Max)
  9. 10131013MiMo-V2.6-Pro
  10. 12001200Qwen3.8 27B (Xhigh)
Datos de Artificial Analysis

Tokens por tarea

tokens de salida por tarea · #11 de 27
  1. 49,2 mil49 milNemotron 3.5 Super VL
  2. 50 mil50 milGPT-6 Luna (Max)
  3. 51,5 mil52 milK2 Horizon 375B A23B
  4. 53,8 mil54 milMuse Wonder Full (Max)
  5. 54,5 mil55 milMuse Wonder (Max)
  6. 55,2 mil55 milKimi K3.1 Pro
  7. 61,6 mil62 milGemini 4 Argon (High)
  8. 64,1 mil64 milStep 5 Preview
  9. 64,3 mil64 milMiMo-V2.6-Pro
  10. 66,8 mil67 milQwen3.8 27B (Xhigh)
Datos de Artificial Analysis

Velocidad y precio

Velocidad

tokens por segundo · #127 de 162
  1. 6060Qwen3.5 397B A17B (Non-reasoning)
  2. 5959Qwen3.5 397B A17B (Reasoning)
  3. 5959GPT-6.1 Sol (Max)
  4. 5858Claude Fable 5.1 (High)
  5. 5858Ling 3.0 Tiny
  6. 5858Kimi K3.1 Pro
  7. 5757Claude Fable 5.1 (Low)
  8. 5757MiMo-V2.6-Pro
  9. 5757GPT-6 Astra (High)
  10. 5656Qwen3.7 Plus
Datos de Artificial Analysis

Latencia

segundos al primer token · #108 de 162
  1. 2,82,8Kimi K2.7 Code
  2. 2,82,8Qwen3.8 Max (0902)
  3. 2,82,8GLM-5.3-Flash
  4. 2,92,9Qwen3.8 2.4T A95B
  5. 33GLM-5.3 (Low)
  6. 3,13,1Kimi K3.1 Pro
  7. 3,13,1MiMo-V2.6-Pro
  8. 3,53,5Apertus 8B Instruct
  9. 3,63,6Mercury 2.5
  10. 3,73,7Ring-2.6-1T
Datos de Artificial Analysis

Precio

$ por 1M tokens (3:1) · #133 de 169
  1. $3,443,44Nova 2.0 Pro Preview (Medium)
  2. $3,443,44Nova 2.0 Pro Preview (Low)
  3. $3,443,44Nova 2.0 Pro Preview (Non-reasoning)
  4. $3,53,5o3
  5. $44Claude Sonnet 5.5 (Max)
  6. $44Kimi K3.1 Pro
  7. $44Gemini 4 Argon (High)
  8. $44Claude Sonnet 5.5 (Xhigh)
  9. $44GPT-6.1 Sol (Max)
  10. $44GPT-6.1 Sol (Xhigh)
Datos de Artificial Analysis

Modelos cercanos

  • Claude Opus 5.5 (Max)Anthropic57,6
  • Claude Opus 5.5 (Xhigh)Anthropic56
  • Claude Sonnet 5.5 (Max)Anthropic56
  • Muse Wonder Full (Max)Meta55,8
  • Claude Opus 5.5 (High)Anthropic53,6
  • Claude Fable 5.1 (Max)Anthropic53,4
  • Claude Fable 5.1 (Xhigh)Anthropic53,2
  • GPT-6 Astra (Max)OpenAI52,7
Datos de Artificial Analysis

Mismo laboratorio

  • Kimi K2.7 CodeKimi25,8
  • Kimi Linear 48B A3B InstructKimi7,3
Datos de Artificial Analysis
HorizonVascend IDEOlympusTalosCursosCommunityBenchmark
DiscordGitHubXHablemosAvisos legalesPrivacidadCookiesUso de datosSeguridad
© 2026 Vascend Technologies · VAT IT04911550616
ascend