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

HyperNova 60B 2605 (High, Based on gpt-oss-120b)

  • Multiverse Computing
  • Release · May 26, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#147 of 266
  1. 12.513MiniCPM5-2B
  2. 12.312Mercury 2.5
  3. 12.212JT-MINI
  4. 11.812Nova 2.0 Lite (Low)
  5. 11.812Magistral Medium 1.2
  6. 11.712HyperNova 60B 2605 (High, Based on gpt-oss-120b)
  7. 11.712Nemotron Cascade 2 30B A3B
  8. 11.612gpt-oss-120b (High)
  9. 11.512K2 Think V2
  10. 11.512LongCat Flash Lite
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy15.2%#170
  • Humanity's Last Exam16.6%#112
  • Terminal-Bench 4.00%#96
  • AA-LCR36.3%#166
  • CritPt0.9%#106
  • GPQA Diamond73.3%#105
  • IFBench66.5%#25
  • τ²-Bench Telecom63.2%#56
  • τ²-Bench Banking5.6%#84
  • Terminal-Bench Hard23.5%#31
HyperNova 60B 2605 (High, Based on gpt-oss-120b)Median
Data from Artificial Analysis

Nearby models

  • MiniCPM5-2BOpenBMB12.5
  • Mercury 2.5Inception12.3
  • JT-MINIChina Mobile12.2
  • Nova 2.0 Lite (Low)Amazon11.8
  • Magistral Medium 1.2Mistral11.8
  • Nemotron Cascade 2 30B A3BNVIDIA11.7
  • gpt-oss-120b (High)OpenAI11.6
  • K2 Think V2Institute of Foundation Models11.5
Data from Artificial Analysis

Same lab

  • Quasar 438B (Max, Based on GLM-5.2)Multiverse Computing26.7
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend