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

LFM2.5-2.6B

  • Liquid AI
  • Release · August 4, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#191 of 266
  1. 8.88.8Sarvam 105B (High)
  2. 8.78.7MiniCPM5-1B (Non-reasoning)
  3. 8.78.7Nova 2.0 Lite (Non-reasoning)
  4. 8.68.6Magistral Small 1.2
  5. 8.48.4Nanbeige4.1-3B
  6. 8.48.4LFM2.5-2.6B
  7. 8.28.2EXAONE 4.0 32B (Reasoning)
  8. 8.28.2Nova 2.0 Omni (Non-reasoning)
  9. 7.97.9Hermes 4 - Llama-3.1 70B (Reasoning)
  10. 7.87.8Falcon-H1R-7B
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy4.4%#253
  • Humanity's Last Exam6.2%#187
  • AA-LCR5.7%#226
  • SciCode14.4%#139
  • CritPt0%#143
  • GPQA Diamond55.8%#156
  • τ²-Bench Banking7.2%#74
LFM2.5-2.6BMedian
Data from Artificial Analysis

Nearby models

  • Sarvam 105B (High)Sarvam8.8
  • MiniCPM5-1B (Non-reasoning)OpenBMB8.7
  • Nova 2.0 Lite (Non-reasoning)Amazon8.7
  • Magistral Small 1.2Mistral8.6
  • Nanbeige4.1-3BNanbeige8.4
  • EXAONE 4.0 32B (Reasoning)LG AI Research8.2
  • Nova 2.0 Omni (Non-reasoning)Amazon8.2
  • Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research7.9
Data from Artificial Analysis

Same lab

  • LFM2.5-8B-A1BLiquid AI7.2
  • LFM2 24B A2BLiquid AI5.9
  • LFM2.5-1.2B-ThinkingLiquid AI5.2
  • LFM2.5-1.2B-InstructLiquid AI5.2
  • LFM2.5-VL-1.6BLiquid AI4.8
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend