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

Granite 4.2 3B

  • IBM
  • Release · August 25, 2026
  • Open weights
  • Reasoning

Intelligence

Intelligence Index

#178 of 266
  1. 9.49.4Gemma 4 12B (Non-reasoning)
  2. 9.39.3Llama 4 Maverick
  3. 9.29.2Qwen3 Coder Next
  4. 9.29.2Motif-2-12.7B-Reasoning
  5. 9.29.2Nova Premier
  6. 9.19.1Granite 4.2 3B
  7. 99gpt-oss-20b (High)
  8. 99Mistral Small 4 (Non-reasoning)
  9. 99K2-V2 (Medium)
  10. 99Tri-21B-Think
Data from Artificial Analysis

Evaluations

  • AA-Omniscience Accuracy9.2%#227
  • Humanity's Last Exam6.6%#179
  • Terminal-Bench 4.00%#96
  • AA-LCR24.3%#184
  • SciCode25.4%#129
  • CritPt0%#143
  • GPQA Diamond55.9%#155
  • τ²-Bench Banking5.6%#84
Granite 4.2 3BMedian
Data from Artificial Analysis

Speed and price

Speed

tokens per second · #143 of 162
  1. 5353GPT-6.1 Sol (Xhigh)
  2. 5353GPT-6 Astra (Low)
  3. 5252GPT-6.1 Sol (Low)
  4. 5252Llama 4 Maverick
  5. 5252Qwen3.8-Flash-Next
  6. 5151Granite 4.2 3B
  7. 5151GPT-6.1 Sol (Medium)
  8. 5050Qwen3.8 27B (Non-reasoning)
  9. 4949Qwen3.8 27B (Low)
  10. 4848Qwen3.8 27B (Medium)
Data from Artificial Analysis

Latency

seconds to first token · #9 of 162
  1. 0.60.6Nemotron 3.5 Lightning
  2. 0.60.6NVIDIA Nemotron Nano 9B V2 (Reasoning)
  3. 0.70.7Ministral 3 3B
  4. 0.70.7Beam 1.0
  5. 0.70.7Nemotron 3.5 Super VL
  6. 0.70.7Granite 4.2 3B
  7. 0.80.8GPT-5.6 Terra (Non-reasoning)
  8. 0.80.8gpt-oss-120b (Low)
  9. 0.80.8Qwen3 Next 80B A3B Instruct
  10. 0.80.8gpt-oss-20b (High)
Data from Artificial Analysis

Price

$ per 1M tokens (3:1) · #9 of 169
  1. $00Ling 3.0 Tiny
  2. $00North Mini Code
  3. $00Tiny Aya Global
  4. $0.040.04Gemma 4 E4B (Reasoning)
  5. $0.040.04Gemma 4 E4B (Non-reasoning)
  6. $0.050.05Granite 4.2 3B
  7. $0.060.06Qwen3.5 4B (Reasoning)
  8. $0.060.06Qwen3.5 4B (Non-reasoning)
  9. $0.060.06Nova Micro
  10. $0.070.07NVIDIA Nemotron Nano 9B V2 (Reasoning)
Data from Artificial Analysis

Nearby models

  • Gemma 4 12B (Non-reasoning)Google9.4
  • Llama 4 MaverickMeta9.3
  • Qwen3 Coder NextAlibaba9.2
  • Motif-2-12.7B-ReasoningMotif Technologies9.2
  • Nova PremierAmazon9.2
  • gpt-oss-20b (High)OpenAI9
  • Mistral Small 4 (Non-reasoning)Mistral9
  • K2-V2 (Medium)Institute of Foundation Models9
Data from Artificial Analysis

Same lab

  • Granite 4.2 30BIBM12.7
  • Granite 4.2 8BIBM11.1
  • Granite 4.0 H SmallIBM6
  • Granite 4.0 H 1BIBM5.2
  • Granite 4.0 MicroIBM5.1
  • Granite 4.0 H 350MIBM4.8
  • Granite 4.0 350MIBM4.8
Data from Artificial Analysis
HorizonVascend IDEOlympusTalosCoursesCommunityBenchmark
DiscordGitHubXLet’s talkLegalPrivacyCookiesData useSecurity
© 2026 Vascend Technologies · VAT IT04911550616
ascend