Granite 4.1 30B vs Llama 3.1 Instruct 405B
Wins 1 of 6 areas
Following instructions
Wins 4 of 6 areas
Coding · Reasoning · Facts · Long documents
Llama 3.1 Instruct 405B is the stronger all-rounder.Granite 4.1 30B is better at following instructions.
Scores updated · 28 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 software02Llama 3.1 Instruct 405B2 of 2 tests
- FactsGetting facts right instead of making them up02Llama 3.1 Instruct 405B2 of 2 tests
- ReasoningHard problems that need careful thinking01Llama 3.1 Instruct 405B1 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts01Llama 3.1 Instruct 405B1 of 1 test
- Following instructionsDoing exactly what it is asked10Granite 4.1 30B1 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
Agents, long documents and following instructions rest on a single test each.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Llama 3.1 Instruct 405B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+42.6points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+9.2points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+4.5points ahead
Where Granite 4.1 30B pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+5.4points ahead
Every test, side by side
All 28 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingLlama 3.1 Instruct 405B
- Terminal-Bench HardLlama 3.1 Instruct 405B by 4.52.36.8+4.5
- SciCodeLlama 3.1 Instruct 405B by 4.125.829.9+4.1
ReasoningLlama 3.1 Instruct 405B
- GPQA DiamondLlama 3.1 Instruct 405B by 3.448.151.5+3.4
- Humanity's Last Examtie4.14tie
- CritPttie00tie
FactsLlama 3.1 Instruct 405B
- AA-Omniscience · Non-hallucinationLlama 3.1 Instruct 405B by 42.6547.6+42.6
- AA-Omniscience · AccuracyLlama 3.1 Instruct 405B by 9.213.923.2+9.2
Long documentsLlama 3.1 Instruct 405B
- AA-LCRLlama 3.1 Instruct 405B by 1.623.725.3+1.6
Following instructionsGranite 4.1 30B
- IFBenchGranite 4.1 30B by 5.444.439+5.4
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.
- AA-OmniscienceLlama 3.1 Instruct 405B by 50.7-67.8-17.1+50.7
- τ²-Bench Telecom (AA run)Granite 4.1 30B by 23.142.119+23.1
- MGSMLlama 3.1 Instruct 405B by 20.571.191.6+20.5
- MBPPGranite 4.1 30B by 17.185.568.4+17.1
- AlpacaEval-2.0Granite 4.1 30B by 16.956.239.3+16.9
- SimpleQALlama 3.1 Instruct 405B by 10.36.817.1+10.3
- MMLU-ProLlama 3.1 Instruct 405B by 9.264.173.3+9.2
- MMLULlama 3.1 Instruct 405B by 8.480.288.6+8.4
- Artificial Analysis Coding IndexLlama 3.1 Instruct 405B by 4.110.414.5+4.1
- CRUXEval-O (output prediction)Llama 3.1 Instruct 405B by 4.155.859.9+4.1
- IFEval (avg)Granite 4.1 30B by 3.389.786.4+3.3
- GSM8KLlama 3.1 Instruct 405B by 2.694.296.8+2.6
- Arena HardGranite 4.1 30B by 1.77169.3+1.7
- BBHtie83.782.9tie
- HumanEvaltie88.489tie
- MBPP+ (EvalPlus-augmented)tie73.573tie
- AA Intelligencetie7.47.3tie
- MMMLUtie73.773.8tie
Questions people ask
Which is better, Granite 4.1 30B or Llama 3.1 Instruct 405B?
Llama 3.1 Instruct 405B wins four of the six areas where both have results: coding, reasoning, facts and long documents. Granite 4.1 30B wins following instructions. They are level on agents.
Which is better for coding?
Llama 3.1 Instruct 405B. It wins 2 of the 2 coding tests both models report; Granite 4.1 30B wins none.
How do you compare the two?
We use the 28 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 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.