Vascend
HorizonVascend IDEVascend CLICommunityBenchmarkContact
Sign inJoin Discord
HomeHorizonIDECLISign in
ChangelogIntelligenceCost per taskCoding Agent IndexCyber IndexDecision IndexBenchmarksSpeedLatencyPriceLeaderboard
Benchmark

Mistral Large 4 Preview

  • Mistral
  • Release · October 6, 2026
  • Proprietary
  • Reasoning

Intelligence

Intelligence Index

#49 of 266
  1. 40.841Claude Sonnet 5.5 (Medium)
  2. 39.940Qwen3.8 2.4T A95B
  3. 39.840Gemini 3.8 Flash (Medium)
  4. 39.840Qwen3.8-Flash-Next
  5. 38.939Muse Spark Open (Xhigh)
  6. 38.438Mistral Large 4 Preview
  7. 38.138GPT-6 Luna (Max)
  8. 3838GPT-5.6 Terra (Xhigh)
  9. 37.938MiMo-V2.6-Flash
  10. 37.838Claude Haiku 5.5 (High)
Data from Artificial Analysis

Evaluations

  • AA-Briefcase1,391#20
  • AA-Omniscience Accuracy25.8%#88
  • AA-Omniscience Non-Hallucination58.1%#14
  • AutomationBench-AA59.9%#14
  • GDPval-AA1,422#21
  • Humanity's Last Exam35%#65
  • GDP.pdf18.6%#17
  • Terminal-Bench 4.026.8%#41
  • AA-LCR81.3%#34
  • SciCode54.2%#38
  • CritPt10.6%#67
  • Terminal-Bench-Science4.8%#17
  • FrontierCode41.5%#15
  • Harvey LAB-AA3.1%#10
  • AA-3D870#15
  • MLCR-AA7.8%#23
  • MMMU-Pro76.4%#42
  • PersonalityBench: Expressiveness84.8#6
  • PersonalityBench: Organization87.3#6
  • PersonalityBench: Transparency34#21
  • PersonalityBench: Slop61.9#8
  • PersonalityBench: Informativeness64.5#12
  • PersonalityBench: Warmth77.2#7
  • PersonalityBench: Conviction34#20
  • PersonalityBench: Imagination59.4#12
Mistral Large 4 PreviewMedian
Data from Artificial Analysis

Cost per task

Cost per task

$ per Intelligence Index task · #13 of 24
  1. $0.280.28GLM-5.3-Flash
  2. $0.720.72GPT-6.1 Sol (Max)
  3. $0.840.84Muse Spark Open (Xhigh)
  4. $1.011.01Qwen3.8 27B (Xhigh)
  5. $1.031.03Step 5 Preview
  6. $1.131.13Mistral Large 4 Preview
  7. $1.241.24Gemini 3.8 Flash (High)
  8. $1.991.99Gemini 4 Argon (High)
  9. $2.012.01GLM-5.3 (Max)
  10. $2.112.11Kimi K3.1 Pro
Data from Artificial Analysis

Time per task

seconds per task · #14 of 26
  1. 396396DeepSeek V4.1 Flash 1017 (Max)
  2. 412412GPT-6 Astra (Max)
  3. 428428Claude Haiku 5.5 (Max)
  4. 498498Gemini 3.8 Flash (High)
  5. 507507MiniMax-M3.1-Flash
  6. 578578Mistral Large 4 Preview
  7. 604604GLM-5.3 (Max)
  8. 628628GPT-6.1 Sol (Max)
  9. 723723Gemini 4 Argon (High)
  10. 736736Claude Fable 5.1 (Max)
Data from Artificial Analysis

Tokens per task

output tokens per task · #17 of 27
  1. 61.6K62KGemini 4 Argon (High)
  2. 64.1K64KStep 5 Preview
  3. 64.3K64KMiMo-V2.6-Pro
  4. 66.8K67KQwen3.8 27B (Xhigh)
  5. 68.7K69KGLM-5.3-Flash
  6. 69.5K70KMistral Large 4 Preview
  7. 71K71KGemini 3.8 Flash (High)
  8. 71.1K71KGLM-5.3 (Max)
  9. 78.1K78KClaude Fable 5.1 (Max)
  10. 78.6K79KMiniMax-M3.1-Flash
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #80 of 162
  1. 114114GPT-5.6 Terra (Medium)
  2. 114114Gemma 4 12B (Reasoning)
  3. 114114Agnes 2.5 Pro Beta
  4. 113113MiniMax-M3.1-Flash
  5. 112112Solar Mini 4
  6. 109109Mistral Large 4 Preview
  7. 108108Claude Sonnet 5.5 (Xhigh)
  8. 108108Gemma 4 26B A4B (Non-reasoning)
  9. 106106Claude Sonnet 5.5 (Medium)
  10. 106106Claude Sonnet 5.5 (High)
Data from Artificial Analysis

Latency

seconds to first token · #68 of 162
  1. 1.61.6Sarvam 105B (High)
  2. 1.71.7Claude Fable 5.1 (Low)
  3. 1.81.8DeepSeek V4 Pro 0813 (Max)
  4. 1.81.8Llama 3.3 Instruct 70B
  5. 1.81.8DeepSeek V4 Pro 0813 (Non-reasoning)
  6. 1.81.8Mistral Large 4 Preview
  7. 1.81.8Inkling (Xhigh)
  8. 1.81.8Qwen3.5 Omni Flash
  9. 1.81.8Command A
  10. 1.81.8Ling 3.1 Flash
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #119 of 169
  1. $1.711.71Kimi K2.7 Code
  2. $1.761.76Inkling (Xhigh)
  3. $1.981.98DeepSeek V4 Pro 0813 (Max)
  4. $1.981.98DeepSeek V4 Pro 0813 (Non-reasoning)
  5. $22Muse Spark Open (Xhigh)
  6. $2.072.07Mistral Large 4 Preview
  7. $2.152.15GLM-5.3 (Max)
  8. $2.152.15GLM-5.3 (Low)
  9. $33Grok 4.7 (Xhigh)
  10. $33Grok 4.7 (High)
Data from Artificial Analysis

Nearby models

  • Claude Sonnet 5.5 (Medium)Anthropic40.8
  • Qwen3.8 2.4T A95BAlibaba39.9
  • Gemini 3.8 Flash (Medium)Google39.8
  • Qwen3.8-Flash-NextAlibaba39.8
  • Muse Spark Open (Xhigh)Meta38.9
  • GPT-6 Luna (Max)OpenAI38.1
  • GPT-5.6 Terra (Xhigh)OpenAI38
  • MiMo-V2.6-FlashXiaomi37.9
Data from Artificial Analysis

Same lab

  • Mistral Medium 3.5Mistral14.2
  • Magistral Medium 1.2Mistral11.8
  • Mistral Small 4 (Reasoning)Mistral11.3
  • Mistral Small 4 (Non-reasoning)Mistral9
  • Magistral Small 1.2Mistral8.6
  • Ministral 3 14BMistral6
  • Ministral 3 8BMistral5.5
  • Ministral 3 3BMistral4.8
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend