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

Muse Spark Open (Xhigh)

  • Meta
  • Release · October 15, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#48 of 266
  1. 40.941Gemini 3.8 Flash (High)
  2. 40.841Claude Sonnet 5.5 (Medium)
  3. 39.940Qwen3.8 2.4T A95B
  4. 39.840Gemini 3.8 Flash (Medium)
  5. 39.840Qwen3.8-Flash-Next
  6. 38.939Muse Spark Open (Xhigh)
  7. 38.438Mistral Large 4 Preview
  8. 38.138GPT-6 Luna (Max)
  9. 3838GPT-5.6 Terra (Xhigh)
  10. 37.938MiMo-V2.6-Flash
Data from Artificial Analysis

Evaluations

  • AA-Briefcase1,279#23
  • AA-Omniscience Accuracy39.3%#57
  • AA-Omniscience Non-Hallucination75.9%#3
  • AutomationBench-AA49.7%#21
  • GDPval-AA1,401#23
  • Humanity's Last Exam43.9%#34
  • GDP.pdf18.6%#17
  • Terminal-Bench 4.08.1%#61
  • AA-LCR78.3%#65
  • SciCode58.5%#18
  • CritPt14.9%#56
  • Terminal-Bench-Science1.9%#22
  • FrontierCode27.7%#24
  • MLCR-AA7.8%#23
  • MMMU-Pro80.9%#22
  • PersonalityBench: Expressiveness51.8#16
  • PersonalityBench: Organization31.5#22
  • PersonalityBench: Transparency51.8#16
  • PersonalityBench: Slop87.3#2
  • PersonalityBench: Informativeness89.8#4
  • PersonalityBench: Warmth87.3#4
  • PersonalityBench: Conviction23.8#23
  • PersonalityBench: Imagination97.5#1
  • GPQA Diamond89%#34
Muse Spark Open (Xhigh)Median
Data from Artificial Analysis

Cost per task

Cost per task

$ per Intelligence Index task · #10 of 24
  1. $0.220.22MiniMax-M3.1-Flash
  2. $0.230.23Beam 1.0
  3. $0.260.26DeepSeek V4.1 Flash 1017 (Max)
  4. $0.280.28GLM-5.3-Flash
  5. $0.720.72GPT-6.1 Sol (Max)
  6. $0.840.84Muse Spark Open (Xhigh)
  7. $1.011.01Qwen3.8 27B (Xhigh)
  8. $1.031.03Step 5 Preview
  9. $1.131.13Mistral Large 4 Preview
  10. $1.241.24Gemini 3.8 Flash (High)
Data from Artificial Analysis

Time per task

seconds per task · #6 of 26
  1. 5555K2 Horizon 375B A23B
  2. 5757Beam 1.0
  3. 9696Nemotron 3.5 Super VL
  4. 165165Inkling (Xhigh)
  5. 263263Step 5 Preview
  6. 284284Muse Spark Open (Xhigh)
  7. 341341Muse Wonder (Max)
  8. 364364GPT-6 Luna (Max)
  9. 396396DeepSeek V4.1 Flash 1017 (Max)
  10. 412412GPT-6 Astra (Max)
Data from Artificial Analysis

Tokens per task

output tokens per task · #5 of 27
  1. 26.3K26KBeam 1.0
  2. 27.2K27KGPT-6 Astra (Max)
  3. 34.7K35KInkling (Xhigh)
  4. 38.1K38KGPT-6.1 Sol (Max)
  5. 40.4K40KMuse Spark Open (Xhigh)
  6. 49.2K49KNemotron 3.5 Super VL
  7. 50K50KGPT-6 Luna (Max)
  8. 51.5K52KK2 Horizon 375B A23B
  9. 53.8K54KMuse Wonder Full (Max)
  10. 54.5K55KMuse Wonder (Max)
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #46 of 162
  1. 162162Muse Wonder (Max)
  2. 159159Quasar 438B (Max, Based on GLM-5.2)
  3. 158158Nemotron 3 Ultra 550B A55B (Reasoning)
  4. 152152GPT-5.5 Instant (June 2026)
  5. 152152o3
  6. 151151Muse Spark Open (Xhigh)
  7. 147147Qwen3.6 35B A3B (Non-reasoning)
  8. 143143Qwen3.5 122B A10B (Non-reasoning)
  9. 140140GPT-6 Luna (Max)
  10. 137137Gemini 3.8 Flash (High)
Data from Artificial Analysis

Latency

seconds to first token · #15 of 162
  1. 0.80.8GPT-5.6 Terra (Non-reasoning)
  2. 0.80.8gpt-oss-120b (Low)
  3. 0.80.8Qwen3 Next 80B A3B Instruct
  4. 0.80.8gpt-oss-20b (High)
  5. 0.80.8Llama 4 Scout
  6. 0.80.8Muse Spark Open (Xhigh)
  7. 0.80.8gpt-oss-120b (High)
  8. 0.80.8Granite 4.2 8B
  9. 0.80.8Gemma 4 E4B (Non-reasoning)
  10. 0.80.8NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #118 of 169
  1. $1.51.5Hermes 4 - Llama-3.1 405B (Non-reasoning)
  2. $1.711.71Kimi K2.7 Code
  3. $1.761.76Inkling (Xhigh)
  4. $1.981.98DeepSeek V4 Pro 0813 (Max)
  5. $1.981.98DeepSeek V4 Pro 0813 (Non-reasoning)
  6. $22Muse Spark Open (Xhigh)
  7. $2.072.07Mistral Large 4 Preview
  8. $2.152.15GLM-5.3 (Max)
  9. $2.152.15GLM-5.3 (Low)
  10. $33Grok 4.7 (Xhigh)
Data from Artificial Analysis

Nearby models

  • Gemini 3.8 Flash (High)Google40.9
  • Claude Sonnet 5.5 (Medium)Anthropic40.8
  • Qwen3.8 2.4T A95BAlibaba39.9
  • Gemini 3.8 Flash (Medium)Google39.8
  • Qwen3.8-Flash-NextAlibaba39.8
  • Mistral Large 4 PreviewMistral38.4
  • GPT-6 Luna (Max)OpenAI38.1
  • GPT-5.6 Terra (Xhigh)OpenAI38
Data from Artificial Analysis

Same lab

  • Muse Wonder Full (Max)Meta55.8
  • Muse Wonder (Max)Meta52.4
  • Muse Glimmer (High)Meta17.5
  • Llama 4 MaverickMeta9.3
  • Llama 3.3 Instruct 70BMeta7.7
  • Llama 3.1 Instruct 405BMeta7.3
  • Llama 4 ScoutMeta6.5
  • Llama 3.2 Instruct 90B (Vision)Meta6.4
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend