Granite 4.1 8B vs Llama 3.1 Instruct 405B
Wins 1 of 6 areas
Agents
Wins 4 of 6 areas
Coding · Reasoning · Facts · Long documents
Llama 3.1 Instruct 405B is the stronger all-rounder.Granite 4.1 8B is better at agents.
Scores updated · 29 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
- AgentsCarrying out multi-step tasks on its own10Granite 4.1 8B1 of 1 test
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
Agents, 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 Llama 3.1 Instruct 405B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+34.7points ahead
- Reasons across sets of long documentsAA-LCR+12points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+10.9points ahead
Where Granite 4.1 8B pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 29 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 8.121.829.9+8.1
- Terminal-Bench HardLlama 3.1 Instruct 405B by 6.806.8+6.8
ReasoningLlama 3.1 Instruct 405B
- GPQA DiamondLlama 3.1 Instruct 405B by 8.243.351.5+8.2
- Humanity's Last Examtie3.84tie
- CritPttie00tie
FactsLlama 3.1 Instruct 405B
- AA-Omniscience · Non-hallucinationLlama 3.1 Instruct 405B by 34.712.947.6+34.7
- AA-Omniscience · AccuracyLlama 3.1 Instruct 405B by 10.912.323.2+10.9
Long documentsLlama 3.1 Instruct 405B
- AA-LCRLlama 3.1 Instruct 405B by 1213.325.3+12
Other results19 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 47-64.1-17.1+47
- MBPPGranite 4.1 8B by 18.987.368.4+18.9
- MMLU-ProLlama 3.1 Instruct 405B by 17.35673.3+17.3
- MMLULlama 3.1 Instruct 405B by 14.873.888.6+14.8
- CRUXEval-O (output prediction)Llama 3.1 Instruct 405B by 12.347.659.9+12.3
- SimpleQALlama 3.1 Instruct 405B by 12.34.817.1+12.3
- AlpacaEval-2.0Granite 4.1 8B by 10.850.139.3+10.8
- MGSMLlama 3.1 Instruct 405B by 9.382.391.6+9.3
- MMMLULlama 3.1 Instruct 405B by 964.873.8+9
- τ²-Bench Telecom (AA run)Granite 4.1 8B by 8.827.819+8.8
- Artificial Analysis Coding IndexLlama 3.1 Instruct 405B by 59.514.5+5
- AA Agentic IndexGranite 4.1 8B by 4.410.76.3+4.4
- GSM8KLlama 3.1 Instruct 405B by 4.392.596.8+4.3
- HumanEvalLlama 3.1 Instruct 405B by 3.685.489+3.6
- BBHLlama 3.1 Instruct 405B by 2.480.582.9+2.4
- MBPP+ (EvalPlus-augmented)tie73.873tie
- AA Intelligencetie6.67.3tie
- IFEval (avg)tie87.186.4tie
- Arena Hardtie6969.3tie
Questions people ask
Which is better, Granite 4.1 8B 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 8B wins agents. They are level on following instructions.
Which is better for coding?
Llama 3.1 Instruct 405B. It wins 2 of the 2 coding tests both models report; Granite 4.1 8B wins none.
How do you compare the two?
We use the 29 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 19 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.