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

Muse Glimmer (High)

  • Meta
  • Release · August 10, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#116 of 266
  1. 18.719JT-35B-Flash
  2. 18.519GPT-6 Luna (Non-reasoning)
  3. 18.418Qwen3.5 397B A17B (Reasoning)
  4. 18.218Qwen3.6 35B A3B (Reasoning)
  5. 17.718Qwen3.5 122B A10B (Non-reasoning)
  6. 17.518Muse Glimmer (High)
  7. 16.917Doubao Seed Code
  8. 16.717Gemma 4 26B A4B (Reasoning)
  9. 16.617Ring-2.6-1T
  10. 15.616K2 Horizon 3.7B
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy27%#85
  • Humanity's Last Exam22%#97
  • Terminal-Bench 4.00.5%#86
  • AA-LCR83.3%#13
  • SciCode44.9%#88
  • CritPt2.6%#91
  • AA-AnalystAgent20%#21
  • MMMU-Pro74.3%#51
  • GPQA Diamond83.5%#57
  • τ²-Bench Banking23.5%#38
Muse Glimmer (High)Median
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #86 of 162
  1. 108108Claude Sonnet 5.5 (Xhigh)
  2. 108108Gemma 4 26B A4B (Non-reasoning)
  3. 106106Claude Sonnet 5.5 (Medium)
  4. 106106Claude Sonnet 5.5 (High)
  5. 105105Claude Sonnet 5.5 (Low)
  6. 105105Muse Glimmer (High)
  7. 105105DeepSeek V4 Pro 0813 (Max)
  8. 103103MiniMax-M3
  9. 103103Qwen3 Omni 30B A3B (Reasoning)
  10. 100100DeepSeek V4 Pro 0813 (Non-reasoning)
Data from Artificial Analysis

Latency

seconds to first token · #45 of 162
  1. 11Llama 4 Maverick
  2. 11Nova 2.0 Pro Preview (Non-reasoning)
  3. 11Nova 2.0 Lite (Non-reasoning)
  4. 11NVIDIA Nemotron Nano 9B V2 (Non-reasoning)
  5. 1.11.1NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)
  6. 1.11.1Muse Glimmer (High)
  7. 1.11.1Gemma 4 31B (Reasoning)
  8. 1.21.2Step 5 Preview
  9. 1.21.2Nemotron 3 Ultra 550B A55B (Reasoning)
  10. 1.21.2DeepSeek V4.1 Flash 1017 (Max)
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #80 of 169
  1. $0.480.48Qwen3 Coder Next
  2. $0.530.53DeepSeek V4.1 Flash 1017 (Max)
  3. $0.530.53MiniMax-M3
  4. $0.530.53Solar Pro 4
  5. $0.530.53Inkling Small
  6. $0.530.53Muse Glimmer (High)
  7. $0.540.54MiMo-V2.6-Pro
  8. $0.60.6Beam 1.0
  9. $0.640.64Nemotron 3 Nano Omni 30B A3B Reasoning
  10. $0.660.66DeepSeek V4 Flash Vision (Max)
Data from Artificial Analysis

Nearby models

  • JT-35B-FlashChina Mobile18.7
  • GPT-6 Luna (Non-reasoning)OpenAI18.5
  • Qwen3.5 397B A17B (Reasoning)Alibaba18.4
  • Qwen3.6 35B A3B (Reasoning)Alibaba18.2
  • Qwen3.5 122B A10B (Non-reasoning)Alibaba17.7
  • Doubao Seed CodeByteDance Seed16.9
  • Gemma 4 26B A4B (Reasoning)Google16.7
  • Ring-2.6-1TInclusionAI16.6
Data from Artificial Analysis

Same lab

  • Muse Wonder Full (Max)Meta55.8
  • Muse Wonder (Max)Meta52.4
  • Muse Spark Open (Xhigh)Meta38.9
  • 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