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

MiMo-V2.6-Flash

  • Xiaomi
  • Release · September 21, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#52 of 266
  1. 39.840Qwen3.8-Flash-Next
  2. 38.939Muse Spark Open (Xhigh)
  3. 38.438Mistral Large 4 Preview
  4. 38.138GPT-6 Luna (Max)
  5. 3838GPT-5.6 Terra (Xhigh)
  6. 37.938MiMo-V2.6-Flash
  7. 37.838Claude Haiku 5.5 (High)
  8. 3636DeepSeek V4 Pro 0813 (Max)
  9. 35.936Claude Sonnet 5.5 (Low)
  10. 35.235Agnes 2.5 Pro Beta
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy27%#85
  • Humanity's Last Exam35.1%#64
  • Terminal-Bench 4.022.7%#46
  • AA-LCR74.3%#88
  • SciCode51.3%#57
  • CritPt12%#63
  • MMMU-Pro73.1%#57
MiMo-V2.6-FlashMedian
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #132 of 162
  1. 5858Kimi K3.1 Pro
  2. 5757Claude Fable 5.1 (Low)
  3. 5757MiMo-V2.6-Pro
  4. 5757GPT-6 Astra (High)
  5. 5656Qwen3.7 Plus
  6. 5656MiMo-V2.6-Flash
  7. 5555Gemma 4 31B (Non-reasoning)
  8. 5555Claude Fable 5.1 (Medium)
  9. 5454Granite 4.2 8B
  10. 5353GPT-6.1 Sol (High)
Data from Artificial Analysis

Latency

seconds to first token · #113 of 162
  1. 3.13.1Kimi K3.1 Pro
  2. 3.13.1MiMo-V2.6-Pro
  3. 3.53.5Apertus 8B Instruct
  4. 3.63.6Mercury 2.5
  5. 3.73.7Ring-2.6-1T
  6. 3.83.8MiMo-V2.6-Flash
  7. 3.93.9Qwen3.8 27B (Xhigh)
  8. 3.93.9Qwen3.8 27B (Low)
  9. 44Qwen3.8 27B (Non-reasoning)
  10. 44Qwen3.8 27B (Medium)
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #33 of 169
  1. $0.150.15Gemma 4 12B (Non-reasoning)
  2. $0.150.15Ministral 3 8B
  3. $0.150.15Qwen3.5 9B (Reasoning)
  4. $0.160.16Gemma 4 26B A4B (Reasoning)
  5. $0.170.17Quasar 438B (Max, Based on GLM-5.2)
  6. $0.180.18MiMo-V2.6-Flash
  7. $0.180.18Solar Mini 4
  8. $0.190.19Gemma 4 26B A4B (Non-reasoning)
  9. $0.190.19Qwen3.5 9B (Non-reasoning)
  10. $0.20.2Claude Haiku 5.5 (Max)
Data from Artificial Analysis

Nearby models

  • Qwen3.8-Flash-NextAlibaba39.8
  • Muse Spark Open (Xhigh)Meta38.9
  • Mistral Large 4 PreviewMistral38.4
  • GPT-6 Luna (Max)OpenAI38.1
  • GPT-5.6 Terra (Xhigh)OpenAI38
  • Claude Haiku 5.5 (High)Anthropic37.8
  • DeepSeek V4 Pro 0813 (Max)DeepSeek36
  • Claude Sonnet 5.5 (Low)Anthropic35.9
Data from Artificial Analysis

Same lab

  • MiMo-V2.6-ProXiaomi46.3
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend