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

GPT-6 Luna (Medium)

  • OpenAI
  • Proprietary
  • Reasoning

Intelligence

Intelligence Index

#71 of 266
  1. 32.933GPT-6 Luna (High)
  2. 32.533GPT-5.3 Codex (Xhigh)
  3. 30.531K2 Horizon 375B A23B
  4. 30.330Beam 1.0
  5. 30.130GPT-5.6 Terra (Medium)
  6. 29.930GPT-6 Luna (Medium)
  7. 29.730Gemini 3.1 Pro Preview
  8. 29.530Claude Haiku 5.5 (Low)
  9. 29.229MiniMax-M3
  10. 28.228Nex-N2-Pro (Based on Qwen3.5-397B-A17B)
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy43.1%#48
  • Humanity's Last Exam28.3%#88
  • Terminal-Bench 4.02.5%#69
  • AA-LCR78.3%#65
  • SciCode50.9%#59
  • CritPt10.6%#67
  • MMMU-Pro77.9%#38
GPT-6 Luna (Medium)Median
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #97 of 162
  1. 9898GLM-5.3 (Max)
  2. 9696Qwen3 Omni 30B A3B Instruct
  3. 9696Claude Opus 5.5 (Max)
  4. 9494Qwen3 Coder Next
  5. 9393Grok 4.7 (Xhigh)
  6. 9292GPT-6 Luna (Medium)
  7. 9191GLM-5.3 (Low)
  8. 9191Gemma 4 26B A4B (Reasoning)
  9. 8989Llama 3.3 Instruct 70B
  10. 8888Reka Flash 3
Data from Artificial Analysis

Latency

seconds to first token · #102 of 162
  1. 2.52.5GPT-6 Luna (Low)
  2. 2.52.5Phi-4
  3. 2.62.6LongCat 2.0
  4. 2.62.6Ling 3.0 Tiny
  5. 2.82.8Hy3
  6. 2.82.8GPT-6 Luna (Medium)
  7. 2.82.8Kimi K2.7 Code
  8. 2.82.8Qwen3.8 Max (0902)
  9. 2.82.8GLM-5.3-Flash
  10. 2.92.9Qwen3.8 2.4T A95B
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #44 of 169
  1. $0.20.2GPT-6 Luna (Max)
  2. $0.20.2Claude Haiku 5.5 (High)
  3. $0.20.2GPT-6 Luna (Xhigh)
  4. $0.20.2Claude Haiku 5.5 (Medium)
  5. $0.20.2GPT-6 Luna (High)
  6. $0.20.2GPT-6 Luna (Medium)
  7. $0.20.2Claude Haiku 5.5 (Low)
  8. $0.20.2GPT-6 Luna (Low)
  9. $0.20.2GPT-6 Luna (Non-reasoning)
  10. $0.20.2Ministral 3 14B
Data from Artificial Analysis

Nearby models

  • GPT-6 Luna (High)OpenAI32.9
  • GPT-5.3 Codex (Xhigh)OpenAI32.5
  • K2 Horizon 375B A23BInstitute of Foundation Models30.5
  • Beam 1.0Reflection30.3
  • GPT-5.6 Terra (Medium)OpenAI30.1
  • Gemini 3.1 Pro PreviewGoogle29.7
  • Claude Haiku 5.5 (Low)Anthropic29.5
  • MiniMax-M3MiniMax29.2
Data from Artificial Analysis

Same lab

  • GPT-6 Astra (Max)OpenAI52.7
  • GPT-6 Astra (Xhigh)OpenAI52.4
  • GPT-6.1 Sol (Max)OpenAI51.8
  • GPT-6.1 Sol (Xhigh)OpenAI51
  • GPT-6 Astra (High)OpenAI50.9
  • GPT-6.1 Sol (High)OpenAI50.2
  • GPT-6 Astra (Medium)OpenAI49.6
  • GPT-6.1 Sol (Medium)OpenAI47.8
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend