Granite 4.0 1B vs Llama 3.2 Instruct 1B
Wins 2 of 5 areas
Coding · Reasoning
Wins 2 of 5 areas
Facts · Following instructions
The two are evenly matched.Granite 4.0 1B is better at reasoning and coding; Llama 3.2 Instruct 1B at facts and following instructions.
Scores updated · 27 tests both models report · How we compare
Where each one wins
Tests won in each of the five areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking10Granite 4.0 1B1 of 3 tests · 2 ties
- CodingWriting and fixing software10Granite 4.0 1B1 of 2 tests · 1 tie
- FactsGetting facts right instead of making them up01Llama 3.2 Instruct 1B1 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked01Llama 3.2 Instruct 1B1 of 1 test
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
Long documents and following instructions rest on a single test each.
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.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Granite 4.0 1B pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+8.5points ahead
- Code for real scientific research problemsSciCode+7points ahead
Where Llama 3.2 Instruct 1B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+27.2points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+2.3points ahead
Every test, side by side
All 27 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGranite 4.0 1B
- SciCodeGranite 4.0 1B by 78.71.7+7
- Terminal-Bench Hardtie00tie
ReasoningGranite 4.0 1B
- GPQA DiamondGranite 4.0 1B by 8.528.119.6+8.5
- Humanity's Last Examtie4.85.5tie
- CritPttie00tie
FactsLlama 3.2 Instruct 1B
- AA-Omniscience · Non-hallucinationLlama 3.2 Instruct 1B by 27.26.533.7+27.2
- AA-Omniscience · Accuracytie6.27tie
Following instructionsLlama 3.2 Instruct 1B
- IFBenchLlama 3.2 Instruct 1B by 2.320.522.8+2.3
Other results18 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- GSM8KGranite 4.0 1B by 37.773.435.7+37.7
- JHumanEval+Granite 4.0 1B by 33.551.217.7+33.5
- BFCL v3Granite 4.0 1B by 3152.421.4+31
- J-BFCLv3Granite 4.0 1B by 29.550.621.1+29.5
- IFEvalGranite 4.0 1B by 27.279.652.4+27.2
- AA-OmniscienceLlama 3.2 Instruct 1B by 26.9-81.6-54.7+26.9
- τ²-Bench Telecom (AA run)Granite 4.0 1B by 22.822.80+22.8
- MATH-500 (EM)Granite 4.0 1B by 21.444.823.4+21.4
- J-GSM8KGranite 4.0 1B by 17.642.825.2+17.6
- Domain AvgGranite 4.0 1B by 17.338.521.2+17.3
- Avg.Granite 4.0 1B by 15.834.118.3+15.8
- J-MATH500Granite 4.0 1B by 1425.411.4+14
- MMLU-ProGranite 4.0 1B by 12.733.520.8+12.7
- JGPQALlama 3.2 Instruct 1B by 7.924.432.3+7.9
- AA Agentic IndexGranite 4.0 1B by 7.67.60+7.6
- JMMLU-ProXtie15.315.9tie
- AA Intelligencetie54.8tie
- JMMLUtie33.934tie
Questions people ask
Which is better, Granite 4.0 1B or Llama 3.2 Instruct 1B?
Granite 4.0 1B and Llama 3.2 Instruct 1B each win two of the five areas where both have results. Granite 4.0 1B wins coding and reasoning; Llama 3.2 Instruct 1B wins facts and following instructions. They are level on long documents.
Which is better for coding?
Granite 4.0 1B. It wins 1 of the 2 coding tests both models report; Llama 3.2 Instruct 1B wins none, and 1 is a tie.
How do you compare the two?
We use the 27 benchmark tests both models have published scores on. The verdict counts the 9 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.