Granite 4.2 30B vs NVIDIA Nemotron 3 Nano 30B A3B
Wins 4 of 6 areas
Coding · Agents · Long documents · Following instructions
Wins 1 of 6 areas
Reasoning
Granite 4.2 30B is the stronger all-rounder.NVIDIA Nemotron 3 Nano 30B A3B is cheaper and better at reasoning.
Scores updated · 23 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 software40Granite 4.2 30B4 of 4 tests
- AgentsCarrying out multi-step tasks on its own20Granite 4.2 30B2 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts10Granite 4.2 30B1 of 1 test
- Following instructionsDoing exactly what it is asked10Granite 4.2 30B1 of 1 test
- ReasoningHard problems that need careful thinking01NVIDIA Nemotron 3 Nano 30B A3B1 of 3 tests · 2 ties
- FactsGetting facts right instead of making them up11Even1 each
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.
NVIDIA Nemotron 3 Nano 30B A3B costs 69% 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 Granite 4.2 30B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+57.7points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+19.9points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+18.2points ahead
Where NVIDIA Nemotron 3 Nano 30B A3B pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+11.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+7.2points 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.
CodingGranite 4.2 30B
- Terminal-Bench 2.1Granite 4.2 30B by 19.926.66.7+19.9
- SWE-bench VerifiedGranite 4.2 30B by 18.25738.8+18.2
- LiveCodeBench v6Granite 4.2 30B by 7.575.868.3+7.5
- SciCodeGranite 4.2 30B by 7.237.830.6+7.2
AgentsGranite 4.2 30B
- τ-Bench V3 · BankingGranite 4.2 30B by 8.414.46+8.4
- GDPValGranite 4.2 30B by 3.83.80+3.8
- Terminal-Bench 4.0tie00tie
ReasoningNVIDIA Nemotron 3 Nano 30B A3B
- GPQA DiamondNVIDIA Nemotron 3 Nano 30B A3B by 11.364.475.7+11.3
- CritPttie0.30.9tie
- Humanity's Last Examtie11.211.4tie
FactsEven
- AA-Omniscience · Non-hallucinationGranite 4.2 30B by 57.774.416.7+57.7
- AA-Omniscience · AccuracyNVIDIA Nemotron 3 Nano 30B A3B by 7.210.117.3+7.2
Following instructionsGranite 4.2 30B
- IFBenchGranite 4.2 30B by 6.177.271.1+6.1
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.
- AA-OmniscienceGranite 4.2 30B by 38.7-12.9-51.6+38.7
- Artificial Analysis Coding IndexGranite 4.2 30B by 15.529.914.4+15.5
- AIME 2025NVIDIA Nemotron 3 Nano 30B A3B by 1089.299.2+10
- AIME25 no toolsNVIDIA Nemotron 3 Nano 30B A3B by 1089.299.2+10
- BFCLv4Granite 4.2 30B by 7.661.453.8+7.6
- IFBench (prompt)Granite 4.2 30B by 5.777.271.5+5.7
- AA IntelligenceGranite 4.2 30B by 3.812.78.9+3.8
- MMLU-Protie77.678.3tie
- Arena-Hard-V2tie67.967.7tie
Questions people ask
Which is better, Granite 4.2 30B or NVIDIA Nemotron 3 Nano 30B A3B?
Granite 4.2 30B wins four of the six areas where both have results: coding, agents, long documents and following instructions. NVIDIA Nemotron 3 Nano 30B A3B wins reasoning, and costs 69% less. They are level on facts.
Which is better for coding?
Granite 4.2 30B. It wins 4 of the 4 coding tests both models report; NVIDIA Nemotron 3 Nano 30B A3B wins none.
Which is cheaper?
Granite 4.2 30B costs $0.16 per million input tokens and $0.65 per million output tokens; NVIDIA Nemotron 3 Nano 30B A3B costs $0.05 and $0.20. That makes NVIDIA Nemotron 3 Nano 30B A3B about 69% 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 14 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.