Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Grok 3 [Beta] is clearly ahead on the aggregate, 38 to 27. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Grok 3 [Beta]'s sharpest advantage is in reasoning, where it averages 47.4 against 30.1. The single biggest benchmark swing on the page is LongBench v2, 53 to 32. Ministral 3 3B does hit back in multilingual, so the answer changes if that is the part of the workload you care about most.
Pick Grok 3 [Beta] if you want the stronger benchmark profile. Ministral 3 3B only becomes the better choice if multilingual is the priority.
Grok 3 [Beta]
35.7
Ministral 3 3B
22.9
Grok 3 [Beta]
16.3
Ministral 3 3B
6.2
Grok 3 [Beta]
43.2
Ministral 3 3B
30.4
Grok 3 [Beta]
47.4
Ministral 3 3B
30.1
Grok 3 [Beta]
34
Ministral 3 3B
24.5
Grok 3 [Beta]
67
Ministral 3 3B
67
Grok 3 [Beta]
58.7
Ministral 3 3B
59.7
Grok 3 [Beta]
49.1
Ministral 3 3B
36
Grok 3 [Beta] is ahead overall, 38 to 27. The biggest single separator in this matchup is LongBench v2, where the scores are 53 and 32.
Grok 3 [Beta] has the edge for knowledge tasks in this comparison, averaging 34 versus 24.5. Inside this category, MMLU is the benchmark that creates the most daylight between them.
Grok 3 [Beta] has the edge for coding in this comparison, averaging 16.3 versus 6.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.
Grok 3 [Beta] has the edge for math in this comparison, averaging 49.1 versus 36. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.
Grok 3 [Beta] has the edge for reasoning in this comparison, averaging 47.4 versus 30.1. Inside this category, LongBench v2 is the benchmark that creates the most daylight between them.
Grok 3 [Beta] has the edge for agentic tasks in this comparison, averaging 35.7 versus 22.9. Inside this category, OSWorld-Verified is the benchmark that creates the most daylight between them.
Grok 3 [Beta] has the edge for multimodal and grounded tasks in this comparison, averaging 43.2 versus 30.4. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
Grok 3 [Beta] and Ministral 3 3B are effectively tied for instruction following here, both landing at 67 on average.
Ministral 3 3B has the edge for multilingual tasks in this comparison, averaging 59.7 versus 58.7. 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.