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

Mi:dm K 2.5 Pro

  • Korea Telecom
  • Release · December 11, 2025
  • Proprietary
  • Reasoning

Intelligence

Intelligence Index

#159 of 266
  1. 11.211Qwen3.5 9B (Reasoning)
  2. 11.211Qwen3 Next 80B A3B (Reasoning)
  3. 11.111Granite 4.2 8B
  4. 11.111Ling 3.0 Tiny
  5. 11.111Nova 2.0 Omni (Low)
  6. 1111Mi:dm K 2.5 Pro
  7. 10.811Trinity Large Thinking
  8. 10.811G9v3-3B
  9. 10.811Qwen3.5 4B (Non-reasoning)
  10. 10.611INTELLECT-3 (Based on GLM-4.5-Air)
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy17.9%#139
  • Humanity's Last Exam8.1%#168
  • AA-LCR9%#217
  • CritPt0%#143
  • GPQA Diamond70.1%#116
  • IFBench49.3%#64
  • τ²-Bench Telecom86.6%#22
  • Terminal-Bench Hard2.3%#100
Mi:dm K 2.5 ProMedian
Data from Artificial Analysis

Nearby models

  • Qwen3.5 9B (Reasoning)Alibaba11.2
  • Qwen3 Next 80B A3B (Reasoning)Alibaba11.2
  • Granite 4.2 8BIBM11.1
  • Ling 3.0 TinyInclusionAI11.1
  • Nova 2.0 Omni (Low)Amazon11.1
  • Trinity Large ThinkingArcee AI10.8
  • G9v3-3BAI9Stars10.8
  • Qwen3.5 4B (Non-reasoning)Alibaba10.8
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend