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

Granite 4.0 H Small

  • IBM
  • Release · September 22, 2025
  • Open weights
  • Non-reasoning

Intelligence

Intelligence Index

#233 of 266
  1. 6.16.1DeepHermes 3 - Mistral 24B Preview (Non-reasoning)
  2. 6.16.1Jamba 1.7 Large
  3. 6.16.1Qwen3.5 0.8B (Reasoning)
  4. 66Ministral 3 14B
  5. 66Qwen3 Omni 30B A3B Instruct
  6. 66Granite 4.0 H Small
  7. 5.95.9Nova Micro
  8. 5.95.9Phi-4
  9. 5.95.9LFM2 24B A2B
  10. 5.85.8NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy14.3%#184
  • Humanity's Last Exam3.8%#248
  • AA-LCR11.3%#213
  • CritPt0%#143
  • GPQA Diamond41.6%#183
  • IFBench31.5%#120
  • τ²-Bench Telecom17.3%#117
  • Terminal-Bench Hard2.3%#100
Granite 4.0 H SmallMedian
Data from Artificial Analysis

Nearby models

  • DeepHermes 3 - Mistral 24B Preview (Non-reasoning)Nous Research6.1
  • Jamba 1.7 LargeAI21 Labs6.1
  • Qwen3.5 0.8B (Reasoning)Alibaba6.1
  • Ministral 3 14BMistral6
  • Qwen3 Omni 30B A3B InstructAlibaba6
  • Nova MicroAmazon5.9
  • Phi-4Microsoft5.9
  • LFM2 24B A2BLiquid AI5.9
Data from Artificial Analysis

Same lab

  • Granite 4.2 30BIBM12.7
  • Granite 4.2 8BIBM11.1
  • Granite 4.2 3BIBM9.1
  • Granite 4.0 H 1BIBM5.2
  • Granite 4.0 MicroIBM5.1
  • Granite 4.0 H 350MIBM4.8
  • Granite 4.0 350MIBM4.8
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend