Granite 4.1 3B vs Llama 3.1 Instruct 405B
Wins 0 of 6 areas
—
Wins 5 of 6 areas
Coding · Reasoning · Facts · Long documents · Following instructions
Llama 3.1 Instruct 405B is the stronger all-rounder.
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 asked01Llama 3.1 Instruct 405B1 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+42.7points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+20.1points ahead
- Reasons across sets of long documentsAA-LCR+19.3points ahead
Where Granite 4.1 3B pulls ahead
No clear win on a test scored out of 100.
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
- SciCodeLlama 3.1 Instruct 405B by 1811.929.9+18
- Terminal-Bench HardLlama 3.1 Instruct 405B by 4.52.36.8+4.5
ReasoningLlama 3.1 Instruct 405B
- GPQA DiamondLlama 3.1 Instruct 405B by 20.131.451.5+20.1
- Humanity's Last Examtie3.44tie
- CritPttie00tie
FactsLlama 3.1 Instruct 405B
- AA-Omniscience · Non-hallucinationLlama 3.1 Instruct 405B by 42.74.947.6+42.7
- AA-Omniscience · AccuracyLlama 3.1 Instruct 405B by 13.99.323.2+13.9
Long documentsLlama 3.1 Instruct 405B
- AA-LCRLlama 3.1 Instruct 405B by 19.3625.3+19.3
Following instructionsLlama 3.1 Instruct 405B
- IFBenchLlama 3.1 Instruct 405B by 5.333.739+5.3
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 59.9-77-17.1+59.9
- Arena HardLlama 3.1 Instruct 405B by 31.537.869.3+31.5
- MMLU-ProLlama 3.1 Instruct 405B by 23.549.873.3+23.5
- MGSMLlama 3.1 Instruct 405B by 21.67091.6+21.6
- MMLULlama 3.1 Instruct 405B by 21.66788.6+21.6
- CRUXEval-O (output prediction)Llama 3.1 Instruct 405B by 19.140.859.9+19.1
- MMMLULlama 3.1 Instruct 405B by 16.257.673.8+16.2
- SimpleQALlama 3.1 Instruct 405B by 13.43.717.1+13.4
- MBPP+ (EvalPlus-augmented)Llama 3.1 Instruct 405B by 10.862.273+10.8
- GSM8KLlama 3.1 Instruct 405B by 9.986.996.8+9.9
- Artificial Analysis Coding IndexLlama 3.1 Instruct 405B by 9.84.714.5+9.8
- HumanEvalLlama 3.1 Instruct 405B by 7.381.789+7.3
- BBHLlama 3.1 Instruct 405B by 7.175.882.9+7.1
- IFEval (avg)Llama 3.1 Instruct 405B by 4.182.386.4+4.1
- MBPPGranite 4.1 3B by 2.871.268.4+2.8
- AA IntelligenceLlama 3.1 Instruct 405B by 1.45.97.3+1.4
- AlpacaEval-2.0tie38.639.3tie
- τ²-Bench Telecom (AA run)tie19.619tie
Questions people ask
Which is better, Granite 4.1 3B or Llama 3.1 Instruct 405B?
Llama 3.1 Instruct 405B wins five of the six areas where both have results: coding, reasoning, facts, long documents and following instructions. Granite 4.1 3B wins none. 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 3B 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.