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

DeepSeek V4 Pro 0813 (Non-reasoning)

  • DeepSeek
  • Open weights
  • Non-reasoning

Intelligence

Intelligence Index

#106 of 266
  1. 21.421Qwen3.5 397B A17B (Non-reasoning)
  2. 21.421Nemotron 3.5 Bolt 74B-A9B
  3. 20.821GPT-5.6 Terra (Non-reasoning)
  4. 20.621K2 Horizon 7B
  5. 20.420Qwen3.5 Omni Plus
  6. 20.420DeepSeek V4 Pro 0813 (Non-reasoning)
  7. 20.220o3
  8. 20.220Qwen3.8 27B (Non-reasoning)
  9. 19.720K-EXAONE 2.0 0803
  10. 19.119LongCat 2.0
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy41.5%#52
  • Humanity's Last Exam10.6%#149
  • Terminal-Bench 4.01.5%#72
  • AA-LCR50.7%#145
  • SciCode39.9%#106
  • CritPt0%#143
DeepSeek V4 Pro 0813 (Non-reasoning)Median
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #90 of 162
  1. 105105Claude Sonnet 5.5 (Low)
  2. 105105Muse Glimmer (High)
  3. 105105DeepSeek V4 Pro 0813 (Max)
  4. 103103MiniMax-M3
  5. 103103Qwen3 Omni 30B A3B (Reasoning)
  6. 100100DeepSeek V4 Pro 0813 (Non-reasoning)
  7. 9999GPT-5.3 Codex (Xhigh)
  8. 9898GLM-5.3 (Max)
  9. 9696Qwen3 Omni 30B A3B Instruct
  10. 9696Claude Opus 5.5 (Max)
Data from Artificial Analysis

Latency

seconds to first token · #67 of 162
  1. 1.61.6Inkling Small
  2. 1.61.6Sarvam 105B (High)
  3. 1.71.7Claude Fable 5.1 (Low)
  4. 1.81.8DeepSeek V4 Pro 0813 (Max)
  5. 1.81.8Llama 3.3 Instruct 70B
  6. 1.81.8DeepSeek V4 Pro 0813 (Non-reasoning)
  7. 1.81.8Mistral Large 4 Preview
  8. 1.81.8Inkling (Xhigh)
  9. 1.81.8Qwen3.5 Omni Flash
  10. 1.81.8Command A
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #117 of 169
  1. $1.51.5Hermes 4 - Llama-3.1 405B (Reasoning)
  2. $1.51.5Hermes 4 - Llama-3.1 405B (Non-reasoning)
  3. $1.711.71Kimi K2.7 Code
  4. $1.761.76Inkling (Xhigh)
  5. $1.981.98DeepSeek V4 Pro 0813 (Max)
  6. $1.981.98DeepSeek V4 Pro 0813 (Non-reasoning)
  7. $22Muse Spark Open (Xhigh)
  8. $2.072.07Mistral Large 4 Preview
  9. $2.152.15GLM-5.3 (Max)
  10. $2.152.15GLM-5.3 (Low)
Data from Artificial Analysis

Nearby models

  • Qwen3.5 397B A17B (Non-reasoning)Alibaba21.4
  • Nemotron 3.5 Bolt 74B-A9BNVIDIA21.4
  • GPT-5.6 Terra (Non-reasoning)OpenAI20.8
  • K2 Horizon 7BInstitute of Foundation Models20.6
  • Qwen3.5 Omni PlusAlibaba20.4
  • o3OpenAI20.2
  • Qwen3.8 27B (Non-reasoning)Alibaba20.2
  • K-EXAONE 2.0 0803LG AI Research19.7
Data from Artificial Analysis

Same lab

  • DeepSeek V4.1 Flash 1017 (Max)DeepSeek45.1
  • DeepSeek V4 Pro 0813 (Max)DeepSeek36
  • DeepSeek V4 Flash Vision (Max)DeepSeek34.8
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend