Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Nemotron-4 15B is clearly ahead on the aggregate, 47 to 31. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Nemotron-4 15B's sharpest advantage is in mathematics, where it averages 61.1 against 40.9. The single biggest benchmark swing on the page is HumanEval, 46 to 16.
Ministral 3 3B (Reasoning) is the reasoning model in the pair, while Nemotron-4 15B 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. Ministral 3 3B (Reasoning) gives you the larger context window at 128K, compared with 32K for Nemotron-4 15B.
Pick Nemotron-4 15B if you want the stronger benchmark profile. Ministral 3 3B (Reasoning) only becomes the better choice if you need the larger 128K context window or you want the stronger reasoning-first profile.
Nemotron-4 15B
41.3
Ministral 3 3B (Reasoning)
34
Nemotron-4 15B
26.5
Ministral 3 3B (Reasoning)
7.2
Nemotron-4 15B
49.6
Ministral 3 3B (Reasoning)
30.4
Nemotron-4 15B
53.4
Ministral 3 3B (Reasoning)
35.3
Nemotron-4 15B
42.6
Ministral 3 3B (Reasoning)
25.2
Nemotron-4 15B
79
Ministral 3 3B (Reasoning)
68
Nemotron-4 15B
72.4
Ministral 3 3B (Reasoning)
59.7
Nemotron-4 15B
61.1
Ministral 3 3B (Reasoning)
40.9
Nemotron-4 15B is ahead overall, 47 to 31. The biggest single separator in this matchup is HumanEval, where the scores are 46 and 16.
Nemotron-4 15B has the edge for knowledge tasks in this comparison, averaging 42.6 versus 25.2. Inside this category, MMLU is the benchmark that creates the most daylight between them.
Nemotron-4 15B has the edge for coding in this comparison, averaging 26.5 versus 7.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.
Nemotron-4 15B has the edge for math in this comparison, averaging 61.1 versus 40.9. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.
Nemotron-4 15B has the edge for reasoning in this comparison, averaging 53.4 versus 35.3. Inside this category, SimpleQA is the benchmark that creates the most daylight between them.
Nemotron-4 15B has the edge for agentic tasks in this comparison, averaging 41.3 versus 34. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.
Nemotron-4 15B has the edge for multimodal and grounded tasks in this comparison, averaging 49.6 versus 30.4. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
Nemotron-4 15B has the edge for instruction following in this comparison, averaging 79 versus 68. Inside this category, IFEval is the benchmark that creates the most daylight between them.
Nemotron-4 15B has the edge for multilingual tasks in this comparison, averaging 72.4 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.