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

Apertus 70B Instruct

  • Swiss AI Initiative
  • Release · September 2, 2025
  • Open weights
  • Non-reasoning

Intelligence

Intelligence Index

#256 of 266
  1. 5.25.2Olmo 3 7B Instruct
  2. 5.25.2Granite 4.0 H 1B
  3. 5.25.2Exaone 4.0 1.2B (Non-reasoning)
  4. 5.15.1Gemma 3 270M
  5. 5.15.1DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)
  6. 5.15.1Apertus 70B Instruct
  7. 5.15.1Granite 4.0 Micro
  8. 55Molmo2-8B
  9. 4.84.8Granite 4.0 H 350M
  10. 4.84.8Granite 4.0 350M
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy11.6%#207
  • Humanity's Last Exam5.5%#196
  • AA-LCR0%#229
  • CritPt0%#143
  • GPQA Diamond27.2%#204
  • IFBench25.9%#139
  • τ²-Bench Telecom12.9%#131
  • Terminal-Bench Hard0%#121
Apertus 70B InstructMedian
Data from Artificial Analysis

Nearby models

  • Olmo 3 7B InstructAllen Institute for AI5.2
  • Granite 4.0 H 1BIBM5.2
  • Exaone 4.0 1.2B (Non-reasoning)LG AI Research5.2
  • Gemma 3 270MGoogle5.1
  • DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Nous Research5.1
  • Granite 4.0 MicroIBM5.1
  • Molmo2-8BAllen Institute for AI5
  • Granite 4.0 H 350MIBM4.8
Data from Artificial Analysis

Same lab

  • Apertus 8B InstructSwiss AI Initiative4.8
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend