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

Gemma 4 31B (Reasoning)

  • Google
  • Release · April 2, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#123 of 266
  1. 16.717Gemma 4 26B A4B (Reasoning)
  2. 16.617Ring-2.6-1T
  3. 15.616K2 Horizon 3.7B
  4. 15.616Qwen3.5 122B A10B (Reasoning)
  5. 15.215Qwen3.6 35B A3B (Non-reasoning)
  6. 14.715Gemma 4 31B (Reasoning)
  7. 14.314ERNIE 5.0 Thinking Preview
  8. 14.214Nova 2.0 Pro Preview (Medium)
  9. 14.214Mistral Medium 3.5
  10. 14.214Gemma 4 12B (Reasoning)
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy20%#115
  • Humanity's Last Exam23.6%#93
  • Terminal-Bench 4.00%#96
  • AA-LCR69.7%#108
  • SciCode45.5%#84
  • CritPt1.4%#99
  • MMMU-Pro73.4%#56
  • GPQA Diamond85.7%#44
  • IFBench75.6%#10
  • τ²-Bench Telecom59.9%#60
  • τ²-Bench Banking14.9%#60
  • Terminal-Bench Hard36.4%#13
Gemma 4 31B (Reasoning)Median
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #158 of 162
  1. 4343Hermes 4 - Llama-3.1 405B (Non-reasoning)
  2. 4343Hermes 4 - Llama-3.1 405B (Reasoning)
  3. 4141Qwen3.8 2.4T A95B
  4. 4040Qwen3.8 Max (0902)
  5. 3939Phi-4
  6. 3535Gemma 4 31B (Reasoning)
  7. 3333Command A
  8. 2626Llama 3.2 Instruct 11B (Vision)
  9. 1919Qwen3.5 4B (Reasoning)
  10. 1414Qwen3.5 4B (Non-reasoning)
Data from Artificial Analysis

Latency

seconds to first token · #46 of 162
  1. 11Nova 2.0 Pro Preview (Non-reasoning)
  2. 11Nova 2.0 Lite (Non-reasoning)
  3. 11NVIDIA Nemotron Nano 9B V2 (Non-reasoning)
  4. 1.11.1NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)
  5. 1.11.1Muse Glimmer (High)
  6. 1.11.1Gemma 4 31B (Reasoning)
  7. 1.21.2Step 5 Preview
  8. 1.21.2Nemotron 3 Ultra 550B A55B (Reasoning)
  9. 1.21.2DeepSeek V4.1 Flash 1017 (Max)
  10. 1.31.3Quasar 438B (Max, Based on GLM-5.2)
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #2 of 169
  1. $00K2 Horizon 375B A23B
  2. $00Gemma 4 31B (Reasoning)
  3. $00Command A+
  4. $00Ling 3.0 Tiny
  5. $00North Mini Code
  6. $00Tiny Aya Global
  7. $0.040.04Gemma 4 E4B (Reasoning)
  8. $0.040.04Gemma 4 E4B (Non-reasoning)
  9. $0.050.05Granite 4.2 3B
  10. $0.060.06Qwen3.5 4B (Reasoning)
Data from Artificial Analysis

Nearby models

  • Gemma 4 26B A4B (Reasoning)Google16.7
  • Ring-2.6-1TInclusionAI16.6
  • K2 Horizon 3.7BInstitute of Foundation Models15.6
  • Qwen3.5 122B A10B (Reasoning)Alibaba15.6
  • Qwen3.6 35B A3B (Non-reasoning)Alibaba15.2
  • ERNIE 5.0 Thinking PreviewBaidu14.3
  • Nova 2.0 Pro Preview (Medium)Amazon14.2
  • Mistral Medium 3.5Mistral14.2
Data from Artificial Analysis

Same lab

  • Gemini 4 Argon (High)Google52.6
  • Gemini 3.8 Flash (High)Google40.9
  • Gemini 3.8 Flash (Medium)Google39.8
  • Gemini 3.8 Flash (Low)Google33.5
  • Gemini 3.1 Pro PreviewGoogle29.7
  • Gemini 3.5 Flash-LiteGoogle22.2
  • Gemma 4 12B (Reasoning)Google14.2
  • Gemma 4 31B (Non-reasoning)Google13.9
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend