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

Qwen3.5 397B A17B (Reasoning)

  • Alibaba
  • Release · February 16, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#113 of 266
  1. 20.220Qwen3.8 27B (Non-reasoning)
  2. 19.720K-EXAONE 2.0 0803
  3. 19.119LongCat 2.0
  4. 18.719JT-35B-Flash
  5. 18.519GPT-6 Luna (Non-reasoning)
  6. 18.418Qwen3.5 397B A17B (Reasoning)
  7. 18.218Qwen3.6 35B A3B (Reasoning)
  8. 17.718Qwen3.5 122B A10B (Non-reasoning)
  9. 17.518Muse Glimmer (High)
  10. 16.917Doubao Seed Code
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy30.8%#77
  • Humanity's Last Exam29%#85
  • Terminal-Bench 4.00%#96
  • AA-LCR77.3%#73
  • SciCode44.8%#89
  • CritPt1.7%#98
  • MMMU-Pro77.3%#40
  • GPQA Diamond89.3%#31
  • IFBench78.8%#6
  • τ²-Bench Telecom95.6%#2
  • τ²-Bench Banking13.4%#63
  • Terminal-Bench Hard40.9%#11
  • APEX-Agents15.3%#8
Qwen3.5 397B A17B (Reasoning)Median
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #123 of 162
  1. 6464Kimi K2.7 Code
  2. 6363Claude Fable 5.1 (Xhigh)
  3. 6262GPT-6 Astra (Xhigh)
  4. 6161Qwen3.5 9B (Reasoning)
  5. 6060Qwen3.5 397B A17B (Non-reasoning)
  6. 5959Qwen3.5 397B A17B (Reasoning)
  7. 5959GPT-6.1 Sol (Max)
  8. 5858Claude Fable 5.1 (High)
  9. 5858Ling 3.0 Tiny
  10. 5858Kimi K3.1 Pro
Data from Artificial Analysis

Latency

seconds to first token · #84 of 162
  1. 22Qwen3.6 35B A3B (Non-reasoning)
  2. 2.12.1GPT-6.1 Sol (Low)
  3. 2.12.1Qwen3.6 35B A3B (Reasoning)
  4. 2.12.1Qwen3.8-Flash-Next
  5. 2.22.2Qwen3.5 397B A17B (Non-reasoning)
  6. 2.22.2Qwen3.5 397B A17B (Reasoning)
  7. 2.22.2K2 Horizon 375B A23B
  8. 2.22.2GPT-5.6 Terra (High)
  9. 2.22.2Gemma 4 12B (Reasoning)
  10. 2.32.3Gemma 4 12B (Non-reasoning)
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #106 of 169
  1. $1.131.13Qwen3.8 27B (Low)
  2. $1.131.13Nemotron 3 Ultra 550B A55B (Reasoning)
  3. $1.131.13Qwen3.8 27B (Non-reasoning)
  4. $1.31.3LongCat 2.0
  5. $1.351.35Qwen3.5 397B A17B (Non-reasoning)
  6. $1.351.35Qwen3.5 397B A17B (Reasoning)
  7. $1.431.43Step 5 Preview
  8. $1.51.5Gemini 3.8 Flash (High)
  9. $1.51.5Gemini 3.8 Flash (Medium)
  10. $1.51.5Gemini 3.8 Flash (Low)
Data from Artificial Analysis

Nearby models

  • Qwen3.8 27B (Non-reasoning)Alibaba20.2
  • K-EXAONE 2.0 0803LG AI Research19.7
  • LongCat 2.0LongCat19.1
  • JT-35B-FlashChina Mobile18.7
  • GPT-6 Luna (Non-reasoning)OpenAI18.5
  • Qwen3.6 35B A3B (Reasoning)Alibaba18.2
  • Qwen3.5 122B A10B (Non-reasoning)Alibaba17.7
  • Muse Glimmer (High)Meta17.5
Data from Artificial Analysis

Same lab

  • Qwen3.8 Max (0902)Alibaba45.4
  • Qwen3.8 2.4T A95BAlibaba39.9
  • Qwen3.8-Flash-NextAlibaba39.8
  • Qwen3.8 27B (Xhigh)Alibaba33.7
  • Qwen3.8 27B (Medium)Alibaba27.6
  • Qwen3.8 27B (Low)Alibaba26.2
  • Qwen3.7 PlusAlibaba25.2
  • Qwen3.5 397B A17B (Non-reasoning)Alibaba21.4
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend