Gemma 3 4B vs Llama 3.2 Instruct 1B
Wins 3 of 5 areas
Coding · Reasoning · Following instructions
Wins 1 of 5 areas
Facts
Gemma 3 4B is the stronger all-rounder.Llama 3.2 Instruct 1B is better at facts.
Scores updated · 23 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.
- CodingWriting and fixing software20Gemma 3 4B2 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking10Gemma 3 4B1 of 3 tests · 2 ties
- Following instructionsDoing exactly what it is asked10Gemma 3 4B1 of 1 test
- FactsGetting facts right instead of making them up01Llama 3.2 Instruct 1B1 of 2 tests · 1 tie
- 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.
Gemma 3 4B costs 34% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Gemma 3 4B pulls ahead
- Recent programming contest problemsLiveCodeBench v6+14.8points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+9.5points ahead
- Code for real scientific research problemsSciCode+5.6points ahead
Where Llama 3.2 Instruct 1B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+31.9points ahead
Every test, side by side
All 23 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemma 3 4B
- LiveCodeBench v6Gemma 3 4B by 14.818.94.1+14.8
- SciCodeGemma 3 4B by 5.67.31.7+5.6
- Terminal-Bench Hardtie0.80tie
ReasoningGemma 3 4B
- GPQA DiamondGemma 3 4B by 9.529.119.6+9.5
- Humanity's Last Examtie5.35.5tie
- CritPttie00tie
FactsLlama 3.2 Instruct 1B
- AA-Omniscience · Non-hallucinationLlama 3.2 Instruct 1B by 31.91.833.7+31.9
- AA-Omniscience · Accuracytie7.77tie
Following instructionsGemma 3 4B
- IFBenchGemma 3 4B by 5.528.322.8+5.5
Other results13 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.
- MGSMGemma 3 4B by 58.287.329.1+58.2
- GSM8KGemma 3 4B by 53.589.235.7+53.5
- MATH-500 (EM)Gemma 3 4B by 49.873.223.4+49.8
- HumanEval+Gemma 3 4B by 39.662.823.2+39.6
- IFEvalGemma 3 4B by 37.890.252.4+37.8
- AA-OmniscienceLlama 3.2 Instruct 1B by 28.2-82.9-54.7+28.2
- MMLU-ProGemma 3 4B by 22.843.620.8+22.8
- LCB v5Gemma 3 4B by 15.519.13.6+15.5
- MMMLUGemma 3 4B by 11.950.138.1+11.9
- MMLUGemma 3 4B by 11.858.446.6+11.8
- τ²-Bench Telecom (AA run)Gemma 3 4B by 550+5
- AA Agentic IndexGemma 3 4B by 1.71.70+1.7
- AA Intelligencetie4.84.8tie
Questions people ask
Which is better, Gemma 3 4B or Llama 3.2 Instruct 1B?
Gemma 3 4B 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?
Gemma 3 4B. It wins 2 of the 3 coding tests both models report; Llama 3.2 Instruct 1B wins none, and 1 is a tie.
Which is cheaper?
Gemma 3 4B costs $0.05 per million input tokens and $0.10 per million output tokens; Llama 3.2 Instruct 1B costs $0.03 and $0.20. That makes Gemma 3 4B about 34% cheaper for the same work.
How do you compare the two?
We use the 23 benchmark tests both models have published scores on. The verdict counts the 10 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 13 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.