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

Quasar 438B (Max, Based on GLM-5.2)

  • Multiverse Computing
  • Release · August 10, 2026
  • Proprietary
  • Reasoning

Intelligence

Intelligence Index

#79 of 266
  1. 29.229MiniMax-M3
  2. 28.228Nex-N2-Pro (Based on Qwen3.5-397B-A17B)
  3. 27.628Qwen3.8 27B (Medium)
  4. 27.528GPT-5.6 Terra (Low)
  5. 27.327JT-4.1 Flash 236B A21B (Non-reasoning)
  6. 26.727Quasar 438B (Max, Based on GLM-5.2)
  7. 26.627Solar Pro 4
  8. 26.527Nemotron 3.5 Super VL
  9. 26.426Apodex 1.1
  10. 26.226Qwen3.8 27B (Low)
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy15.5%#167
  • Humanity's Last Exam18.7%#107
  • Terminal-Bench 4.01%#77
  • AA-LCR76.3%#80
  • SciCode48.2%#73
  • CritPt9.4%#71
  • GPQA Diamond73.2%#107
  • τ²-Bench Banking28.7%#33
Quasar 438B (Max, Based on GLM-5.2)Median
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #42 of 162
  1. 169169DeepSeek V4.1 Flash 1017 (Max)
  2. 168168NVIDIA Nemotron Nano 9B V2 (Non-reasoning)
  3. 166166Command A+
  4. 162162Mistral Medium 3.5
  5. 162162Muse Wonder (Max)
  6. 159159Quasar 438B (Max, Based on GLM-5.2)
  7. 158158Nemotron 3 Ultra 550B A55B (Reasoning)
  8. 152152GPT-5.5 Instant (June 2026)
  9. 152152o3
  10. 151151Muse Spark Open (Xhigh)
Data from Artificial Analysis

Latency

seconds to first token · #50 of 162
  1. 1.11.1Muse Glimmer (High)
  2. 1.11.1Gemma 4 31B (Reasoning)
  3. 1.21.2Step 5 Preview
  4. 1.21.2Nemotron 3 Ultra 550B A55B (Reasoning)
  5. 1.21.2DeepSeek V4.1 Flash 1017 (Max)
  6. 1.31.3Quasar 438B (Max, Based on GLM-5.2)
  7. 1.31.3Trinity Large Thinking
  8. 1.31.3MiniMax-M3
  9. 1.31.3Llama 3.2 Instruct 11B (Vision)
  10. 1.41.4MiniMax-M3.1-Flash
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #32 of 169
  1. $0.150.15Gemma 4 12B (Reasoning)
  2. $0.150.15Gemma 4 12B (Non-reasoning)
  3. $0.150.15Ministral 3 8B
  4. $0.150.15Qwen3.5 9B (Reasoning)
  5. $0.160.16Gemma 4 26B A4B (Reasoning)
  6. $0.170.17Quasar 438B (Max, Based on GLM-5.2)
  7. $0.180.18MiMo-V2.6-Flash
  8. $0.180.18Solar Mini 4
  9. $0.190.19Gemma 4 26B A4B (Non-reasoning)
  10. $0.190.19Qwen3.5 9B (Non-reasoning)
Data from Artificial Analysis

Nearby models

  • MiniMax-M3MiniMax29.2
  • Nex-N2-Pro (Based on Qwen3.5-397B-A17B)Nex AGI28.2
  • Qwen3.8 27B (Medium)Alibaba27.6
  • GPT-5.6 Terra (Low)OpenAI27.5
  • JT-4.1 Flash 236B A21B (Non-reasoning)China Mobile27.3
  • Solar Pro 4Upstage26.6
  • Nemotron 3.5 Super VLNVIDIA26.5
  • Apodex 1.1Apodex26.4
Data from Artificial Analysis

Same lab

  • HyperNova 60B 2605 (High, Based on gpt-oss-120b)Multiverse Computing11.7
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend