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

LFM2 24B A2B

  • Liquid AI
  • Release · February 25, 2026
  • Open weights
  • Non-reasoning

Intelligence

Intelligence Index

#236 of 266
  1. 66Ministral 3 14B
  2. 66Qwen3 Omni 30B A3B Instruct
  3. 66Granite 4.0 H Small
  4. 5.95.9Nova Micro
  5. 5.95.9Phi-4
  6. 5.95.9LFM2 24B A2B
  7. 5.85.8NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)
  8. 5.85.8Phi-4 Multimodal Instruct
  9. 5.75.7MiniCPM-V 4.6 1.3B
  10. 5.75.7Jamba Reasoning 3B
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy6.5%#246
  • Humanity's Last Exam4.2%#240
  • AA-LCR0%#229
  • CritPt0%#143
  • GPQA Diamond47.4%#172
  • IFBench45.9%#68
  • τ²-Bench Telecom11.1%#137
  • Terminal-Bench Hard0%#121
LFM2 24B A2BMedian
Data from Artificial Analysis

Nearby models

  • Ministral 3 14BMistral6
  • Qwen3 Omni 30B A3B InstructAlibaba6
  • Granite 4.0 H SmallIBM6
  • Nova MicroAmazon5.9
  • Phi-4Microsoft5.9
  • NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIA5.8
  • Phi-4 Multimodal InstructMicrosoft5.8
  • MiniCPM-V 4.6 1.3BOpenBMB5.7
Data from Artificial Analysis

Same lab

  • LFM2.5-2.6BLiquid AI8.4
  • LFM2.5-8B-A1BLiquid AI7.2
  • 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