Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

Granite 3.3 2B Instruct vs Granite 3.3 8B Instruct

IBM · released

IBM · released

Scores updated · 14 tests both models report · How we compare

What it costs

Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.

Granite 3.3 2B InstructNo current price is tracked.
Granite 3.3 8B Instruct$0.03 to read · $0.25 to write$0.28Price from Artificial Analysis

Every test, side by side

All 14 tests both models report. The winning score is in its model's colour; marks a score checked independently.

Other results14 tests
  • Arena HardGranite 3.3 8B Instruct by 28.728.957.6+28.7
  • AlpacaEval-2.0Granite 3.3 8B Instruct by 19.243.562.7+19.2
  • DROPGranite 3.3 8B Instruct by 15.144.359.4+15.1
  • MATH-500 (EM)Granite 3.3 8B Instruct by 10.958.169+10.9
  • HumanEval+Granite 3.3 8B Instruct by 10.475.786.1+10.4
  • MMLUGranite 3.3 8B Instruct by 9.655.965.5+9.6
  • HumanEvalGranite 3.3 8B Instruct by 9.280.589.7+9.2
  • IFEvalGranite 3.3 8B Instruct by 965.874.8+9
  • GSM8KGranite 3.3 8B Instruct by 8.472.580.9+8.4
  • TruthfulQAGranite 3.3 8B Instruct by 7.95966.9+7.9
  • PopQAGranite 3.3 8B Instruct by 7.818.426.2+7.8
  • BigBenchHardGranite 3.3 8B Instruct by 5.263.969.1+5.2
  • AIME 2024Granite 3.3 8B Instruct by 4.83.38.1+4.8
  • AttaQGranite 3.3 8B Instruct by 187.588.5+1

Questions people ask

How do you compare the two?

We use the 14 benchmark tests both models have published scores on. None of them is in the eight capability areas we count, so this page lists them without a verdict. Each score is the one shown on the model's own page.