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

Gemma 4 E4B (Reasoning)

  • Google
  • Release · April 3, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#183 of 266
  1. 9.19.1Granite 4.2 3B
  2. 99gpt-oss-20b (High)
  3. 99Mistral Small 4 (Non-reasoning)
  4. 99K2-V2 (Medium)
  5. 99Tri-21B-Think
  6. 8.98.9Gemma 4 E4B (Reasoning)
  7. 8.98.9NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)
  8. 8.88.8MiniCPM5-1B (Reasoning)
  9. 8.88.8Sarvam 105B (High)
  10. 8.78.7MiniCPM5-1B (Non-reasoning)
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy8.6%#230
  • Humanity's Last Exam3.8%#248
  • Terminal-Bench 4.00%#96
  • AA-LCR32%#172
  • SciCode24.4%#130
  • CritPt0.6%#112
  • MMMU-Pro51.4%#100
  • GPQA Diamond57.6%#148
  • IFBench44.2%#73
  • τ²-Bench Telecom20.8%#107
  • Terminal-Bench Hard8.3%#64
Gemma 4 E4B (Reasoning)Median
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #150 of 162
  1. 5050Qwen3.8 27B (Non-reasoning)
  2. 4949Qwen3.8 27B (Low)
  3. 4848Qwen3.8 27B (Medium)
  4. 4848Granite 4.2 30B
  5. 4646Phi-4 Mini Instruct
  6. 4545Gemma 4 E4B (Reasoning)
  7. 4545Qwen3.8 27B (Xhigh)
  8. 4545Gemma 4 E4B (Non-reasoning)
  9. 4343Hermes 4 - Llama-3.1 405B (Non-reasoning)
  10. 4343Hermes 4 - Llama-3.1 405B (Reasoning)
Data from Artificial Analysis

Latency

seconds to first token · #30 of 162
  1. 0.90.9Mistral Small 4 (Reasoning)
  2. 0.90.9GPT-6 Luna (Non-reasoning)
  3. 0.90.9GPT-5.5 Instant (June 2026)
  4. 0.90.9Qwen3.5 9B (Non-reasoning)
  5. 0.90.9Qwen3.5 4B (Non-reasoning)
  6. 0.90.9Gemma 4 E4B (Reasoning)
  7. 0.90.9Gemma 4 26B A4B (Reasoning)
  8. 0.90.9Qwen3.5 4B (Reasoning)
  9. 0.90.9Qwen3 Coder Next
  10. 0.90.9Mistral Small 4 (Non-reasoning)
Data from Artificial Analysis

Price

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

Nearby models

  • Granite 4.2 3BIBM9.1
  • gpt-oss-20b (High)OpenAI9
  • Mistral Small 4 (Non-reasoning)Mistral9
  • K2-V2 (Medium)Institute of Foundation Models9
  • Tri-21B-ThinkTrillion Labs9
  • NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA8.9
  • MiniCPM5-1B (Reasoning)OpenBMB8.8
  • Sarvam 105B (High)Sarvam8.8
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 26B A4B (Reasoning)Google16.7
  • Gemma 4 31B (Reasoning)Google14.7
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend