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

Agnes 2.5 Pro Beta

  • Sapiens AI
  • Release · August 26, 2026
  • Proprietary
  • Reasoning

Intelligence

Intelligence Index

#56 of 266
  1. 3838GPT-5.6 Terra (Xhigh)
  2. 37.938MiMo-V2.6-Flash
  3. 37.838Claude Haiku 5.5 (High)
  4. 3636DeepSeek V4 Pro 0813 (Max)
  5. 35.936Claude Sonnet 5.5 (Low)
  6. 35.235Agnes 2.5 Pro Beta
  7. 34.835DeepSeek V4 Flash Vision (Max)
  8. 34.635GPT-6 Luna (Xhigh)
  9. 34.535Claude Haiku 5.5 (Medium)
  10. 34.334GLM-5.3 (Low)
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy16.8%#152
  • Humanity's Last Exam37.5%#57
  • AA-LCR83%#18
  • SciCode48.8%#69
  • CritPt15.7%#53
  • MMMU-Pro75.4%#47
  • GPQA Diamond90.5%#24
  • τ²-Bench Banking35.7%#22
Agnes 2.5 Pro BetaMedian
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #77 of 162
  1. 116116GPT-6 Luna (Low)
  2. 116116GPT-5.6 Terra (Low)
  3. 115115Nova 2.0 Pro Preview (Non-reasoning)
  4. 114114GPT-5.6 Terra (Medium)
  5. 114114Gemma 4 12B (Reasoning)
  6. 114114Agnes 2.5 Pro Beta
  7. 113113MiniMax-M3.1-Flash
  8. 112112Solar Mini 4
  9. 109109Mistral Large 4 Preview
  10. 108108Claude Sonnet 5.5 (Xhigh)
Data from Artificial Analysis

Latency

seconds to first token · #57 of 162
  1. 1.31.3MiniMax-M3
  2. 1.31.3Llama 3.2 Instruct 11B (Vision)
  3. 1.41.4MiniMax-M3.1-Flash
  4. 1.41.4Claude Sonnet 5.5 (Medium)
  5. 1.41.4GPT-5.6 Terra (Low)
  6. 1.41.4Agnes 2.5 Pro Beta
  7. 1.41.4Reka Flash 3
  8. 1.51.5GPT-5.6 Terra (Medium)
  9. 1.51.5Magistral Small 1.2
  10. 1.61.6DeepSeek V4 Flash Vision (Max)
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #26 of 169
  1. $0.110.11Granite 4.2 8B
  2. $0.130.13Apertus 8B Instruct
  3. $0.130.13gpt-oss-20b (Low)
  4. $0.130.13Phi-4 Mini Instruct
  5. $0.140.14Phi-4 Multimodal Instruct
  6. $0.150.15Agnes 2.5 Pro Beta
  7. $0.150.15Gemma 4 12B (Reasoning)
  8. $0.150.15Gemma 4 12B (Non-reasoning)
  9. $0.150.15Ministral 3 8B
  10. $0.150.15Qwen3.5 9B (Reasoning)
Data from Artificial Analysis

Nearby models

  • GPT-5.6 Terra (Xhigh)OpenAI38
  • MiMo-V2.6-FlashXiaomi37.9
  • Claude Haiku 5.5 (High)Anthropic37.8
  • DeepSeek V4 Pro 0813 (Max)DeepSeek36
  • Claude Sonnet 5.5 (Low)Anthropic35.9
  • DeepSeek V4 Flash Vision (Max)DeepSeek34.8
  • GPT-6 Luna (Xhigh)OpenAI34.6
  • Claude Haiku 5.5 (Medium)Anthropic34.5
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend