Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Llama 3.1 405B is clearly ahead on the aggregate, 59 to 31. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Llama 3.1 405B's sharpest advantage is in mathematics, where it averages 74.9 against 40.9. The single biggest benchmark swing on the page is HumanEval, 62 to 16.
Ministral 3 3B (Reasoning) is the reasoning model in the pair, while Llama 3.1 405B is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use.
Pick Llama 3.1 405B if you want the stronger benchmark profile. Ministral 3 3B (Reasoning) only becomes the better choice if you want the stronger reasoning-first profile.
Llama 3.1 405B
53.9
Ministral 3 3B (Reasoning)
34
Llama 3.1 405B
40.6
Ministral 3 3B (Reasoning)
7.2
Llama 3.1 405B
62.3
Ministral 3 3B (Reasoning)
30.4
Llama 3.1 405B
68.3
Ministral 3 3B (Reasoning)
35.3
Llama 3.1 405B
53.2
Ministral 3 3B (Reasoning)
25.2
Llama 3.1 405B
86
Ministral 3 3B (Reasoning)
68
Llama 3.1 405B
80.1
Ministral 3 3B (Reasoning)
59.7
Llama 3.1 405B
74.9
Ministral 3 3B (Reasoning)
40.9
Llama 3.1 405B is ahead overall, 59 to 31. The biggest single separator in this matchup is HumanEval, where the scores are 62 and 16.
Llama 3.1 405B has the edge for knowledge tasks in this comparison, averaging 53.2 versus 25.2. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Llama 3.1 405B has the edge for coding in this comparison, averaging 40.6 versus 7.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.
Llama 3.1 405B has the edge for math in this comparison, averaging 74.9 versus 40.9. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.
Llama 3.1 405B has the edge for reasoning in this comparison, averaging 68.3 versus 35.3. Inside this category, SimpleQA is the benchmark that creates the most daylight between them.
Llama 3.1 405B has the edge for agentic tasks in this comparison, averaging 53.9 versus 34. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.
Llama 3.1 405B has the edge for multimodal and grounded tasks in this comparison, averaging 62.3 versus 30.4. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
Llama 3.1 405B has the edge for instruction following in this comparison, averaging 86 versus 68. Inside this category, IFEval is the benchmark that creates the most daylight between them.
Llama 3.1 405B has the edge for multilingual tasks in this comparison, averaging 80.1 versus 59.7. Inside this category, MGSM is the benchmark that creates the most daylight between them.
Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.
Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.