Gemma 3 4B vs Granite 4.1 30B
Wins 0 of 6 areas
—
Wins 5 of 6 areas
Coding · Agents · Facts · Long documents · Following instructions
Granite 4.1 30B is the stronger all-rounder.
Scores updated · 26 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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 software02Granite 4.1 30B2 of 2 tests
- FactsGetting facts right instead of making them up02Granite 4.1 30B2 of 2 tests
- AgentsCarrying out multi-step tasks on its own01Granite 4.1 30B1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts01Granite 4.1 30B1 of 1 test
- Following instructionsDoing exactly what it is asked01Granite 4.1 30B1 of 1 test
- ReasoningHard problems that need careful thinking11Even1 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 4.1 30B pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+19points ahead
- Code for real scientific research problemsSciCode+18.5points ahead
- Reasons across sets of long documentsAA-LCR+17points ahead
Where Gemma 3 4B pulls ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+1.2points ahead
Every test, side by side
All 26 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGranite 4.1 30B
- SciCodeGranite 4.1 30B by 18.57.325.8+18.5
- Terminal-Bench HardGranite 4.1 30B by 1.50.82.3+1.5
AgentsGranite 4.1 30B
- τ-Bench V3 · BankingGranite 4.1 30B by 3.50.43.9+3.5
- GDPValtie00tie
ReasoningEven
- GPQA DiamondGranite 4.1 30B by 1929.148.1+19
- Humanity's Last ExamGemma 3 4B by 1.25.34.1+1.2
- CritPttie00tie
FactsGranite 4.1 30B
- AA-Omniscience · AccuracyGranite 4.1 30B by 6.37.713.9+6.3
- AA-Omniscience · Non-hallucinationGranite 4.1 30B by 3.21.85+3.2
Following instructionsGranite 4.1 30B
- IFBenchGranite 4.1 30B by 16.128.344.4+16.1
Other results15 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.
- Arena HardGranite 4.1 30B by 39.231.871+39.2
- τ²-Bench Telecom (AA run)Granite 4.1 30B by 37.1542.1+37.1
- MMMLUGranite 4.1 30B by 23.650.173.7+23.6
- HumanEval+Granite 4.1 30B by 22.662.885.4+22.6
- MBPPGranite 4.1 30B by 22.363.285.5+22.3
- MMLUGranite 4.1 30B by 21.858.480.2+21.8
- MMLU-ProGranite 4.1 30B by 20.543.664.1+20.5
- HumanEvalGranite 4.1 30B by 17.171.388.4+17.1
- MGSMGemma 3 4B by 16.287.371.1+16.2
- AA-OmniscienceGranite 4.1 30B by 15.1-82.9-67.8+15.1
- BBHGranite 4.1 30B by 11.572.283.7+11.5
- Artificial Analysis Coding IndexGranite 4.1 30B by 7.72.710.4+7.7
- GSM8KGranite 4.1 30B by 589.294.2+5
- SimpleQAGranite 4.1 30B by 2.846.8+2.8
- AA IntelligenceGranite 4.1 30B by 2.64.87.4+2.6
Questions people ask
Which is better, Gemma 3 4B or Granite 4.1 30B?
Granite 4.1 30B wins five of the six areas where both have results: coding, agents, facts, long documents and following instructions. Gemma 3 4B wins none. They are level on reasoning.
Which is better for coding?
Granite 4.1 30B. It wins 2 of the 2 coding tests both models report; Gemma 3 4B wins none.
How do you compare the two?
We use the 26 benchmark tests both models have published scores on. The verdict counts the 11 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 15 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.