Granite 4.0 H 1B vs Llama 3.2 Instruct 1B
Wins 3 of 5 areas
Coding · Reasoning · Following instructions
Wins 1 of 5 areas
Facts
Granite 4.0 H 1B is the stronger all-rounder.Llama 3.2 Instruct 1B is better at facts.
Scores updated · 18 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 H 1B1 of 3 tests · 2 ties
- CodingWriting and fixing software10Granite 4.0 H 1B1 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked10Granite 4.0 H 1B1 of 1 test
- FactsGetting facts right instead of making them up02Llama 3.2 Instruct 1B2 of 2 tests
- 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 H 1B pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+6.7points ahead
- Code for real scientific research problemsSciCode+6.5points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+3.4points ahead
Where Llama 3.2 Instruct 1B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+15.4points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+1.8points ahead
Every test, side by side
All 18 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGranite 4.0 H 1B
- SciCodeGranite 4.0 H 1B by 6.58.21.7+6.5
- Terminal-Bench Hardtie00tie
ReasoningGranite 4.0 H 1B
- GPQA DiamondGranite 4.0 H 1B by 6.726.319.6+6.7
- Humanity's Last Examtie55.5tie
- CritPttie00tie
FactsLlama 3.2 Instruct 1B
- AA-Omniscience · Non-hallucinationLlama 3.2 Instruct 1B by 15.418.333.7+15.4
- AA-Omniscience · AccuracyLlama 3.2 Instruct 1B by 1.85.27+1.8
Following instructionsGranite 4.0 H 1B
- IFBenchGranite 4.0 H 1B by 3.426.222.8+3.4
Other results9 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 H 1B by 33.969.635.7+33.9
- BFCL v3Granite 4.0 H 1B by 29.350.721.4+29.3
- IFEvalGranite 4.0 H 1B by 27.780.152.4+27.7
- MATH-500 (EM)Granite 4.0 H 1B by 23.847.223.4+23.8
- τ²-Bench Telecom (AA run)Granite 4.0 H 1B by 19.619.60+19.6
- AA-OmniscienceLlama 3.2 Instruct 1B by 17.6-72.3-54.7+17.6
- MMLU-ProGranite 4.0 H 1B by 6.827.620.8+6.8
- AA Agentic IndexGranite 4.0 H 1B by 6.56.50+6.5
- AA Intelligencetie5.24.8tie
Questions people ask
Which is better, Granite 4.0 H 1B or Llama 3.2 Instruct 1B?
Granite 4.0 H 1B wins three of the five areas where both have results: coding, reasoning and following instructions. Llama 3.2 Instruct 1B wins facts. They are level on long documents.
Which is better for coding?
Granite 4.0 H 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 18 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 9 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.