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

Ministral 3 8B

  • Mistral
  • Release · December 2, 2025
  • Open weights
  • Non-reasoning

Intelligence

Intelligence Index

#244 of 266
  1. 5.75.7MiniCPM-V 4.6 1.3B
  2. 5.75.7Jamba Reasoning 3B
  3. 5.65.6Olmo 3 7B Think
  4. 5.65.6Molmo 7B-D
  5. 5.65.6Reka Flash 3
  6. 5.55.5Ministral 3 8B
  7. 5.45.4Qwen3.5 0.8B (Non-reasoning)
  8. 5.45.4Llama 3.2 Instruct 11B (Vision)
  9. 5.35.3Exaone 4.0 1.2B (Reasoning)
  10. 5.25.2LFM2.5-1.2B-Thinking
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy13%#199
  • Humanity's Last Exam4.3%#236
  • Terminal-Bench 4.00%#96
  • AA-LCR25.7%#181
  • SciCode20.7%#136
  • CritPt0%#143
  • MMMU-Pro46%#106
  • GPQA Diamond47.1%#173
  • IFBench29.1%#128
  • τ²-Bench Telecom26.6%#87
  • τ²-Bench Banking3.7%#96
  • Terminal-Bench Hard4.6%#81
Ministral 3 8BMedian
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #67 of 162
  1. 122122GPT-5.6 Terra (Non-reasoning)
  2. 122122Qwen3.5 Omni Plus
  3. 120120GPT-6 Luna (Non-reasoning)
  4. 120120Nova 2.0 Pro Preview (Low)
  5. 120120GPT-6 Luna (High)
  6. 119119Ministral 3 8B
  7. 119119GPT-5.6 Terra (High)
  8. 118118GPT-5.6 Terra (Xhigh)
  9. 118118Llama 4 Scout
  10. 118118Gemma 4 12B (Non-reasoning)
Data from Artificial Analysis

Latency

seconds to first token · #24 of 162
  1. 0.80.8NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)
  2. 0.80.8Nova Micro
  3. 0.80.8Granite 4.2 30B
  4. 0.80.8Nemotron 3 Nano Omni 30B A3B Reasoning
  5. 0.90.9Phi-4 Mini Instruct
  6. 0.90.9Ministral 3 8B
  7. 0.90.9Mistral Small 4 (Reasoning)
  8. 0.90.9GPT-6 Luna (Non-reasoning)
  9. 0.90.9GPT-5.5 Instant (June 2026)
  10. 0.90.9Qwen3.5 9B (Non-reasoning)
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #29 of 169
  1. $0.130.13Phi-4 Mini Instruct
  2. $0.140.14Phi-4 Multimodal Instruct
  3. $0.150.15Agnes 2.5 Pro Beta
  4. $0.150.15Gemma 4 12B (Reasoning)
  5. $0.150.15Gemma 4 12B (Non-reasoning)
  6. $0.150.15Ministral 3 8B
  7. $0.150.15Qwen3.5 9B (Reasoning)
  8. $0.160.16Gemma 4 26B A4B (Reasoning)
  9. $0.170.17Quasar 438B (Max, Based on GLM-5.2)
  10. $0.180.18MiMo-V2.6-Flash
Data from Artificial Analysis

Nearby models

  • MiniCPM-V 4.6 1.3BOpenBMB5.7
  • Jamba Reasoning 3BAI21 Labs5.7
  • Olmo 3 7B ThinkAllen Institute for AI5.6
  • Molmo 7B-DAllen Institute for AI5.6
  • Reka Flash 3Reka AI5.6
  • Qwen3.5 0.8B (Non-reasoning)Alibaba5.4
  • Llama 3.2 Instruct 11B (Vision)Meta5.4
  • Exaone 4.0 1.2B (Reasoning)LG AI Research5.3
Data from Artificial Analysis

Same lab

  • Mistral Large 4 PreviewMistral38.4
  • Mistral Medium 3.5Mistral14.2
  • Magistral Medium 1.2Mistral11.8
  • Mistral Small 4 (Reasoning)Mistral11.3
  • Mistral Small 4 (Non-reasoning)Mistral9
  • Magistral Small 1.2Mistral8.6
  • Ministral 3 14BMistral6
  • Ministral 3 3BMistral4.8
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend