Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
GPT-4o mini is clearly ahead on the aggregate, 52 to 31. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GPT-4o mini's sharpest advantage is in coding, where it averages 65 against 7.2. The single biggest benchmark swing on the page is HumanEval, 87.2 to 16.
GPT-4o mini is also the more expensive model on tokens at $0.15 input / $0.60 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Ministral 3 3B (Reasoning). That is roughly Infinityx on output cost alone. Ministral 3 3B (Reasoning) is the reasoning model in the pair, while GPT-4o mini 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 GPT-4o mini if you want the stronger benchmark profile. Ministral 3 3B (Reasoning) only becomes the better choice if you want the cheaper token bill or you want the stronger reasoning-first profile.
GPT-4o mini
50.9
Ministral 3 3B (Reasoning)
34
GPT-4o mini
65
Ministral 3 3B (Reasoning)
7.2
GPT-4o mini
60.2
Ministral 3 3B (Reasoning)
30.4
GPT-4o mini
49.4
Ministral 3 3B (Reasoning)
35.3
GPT-4o mini
62
Ministral 3 3B (Reasoning)
25.2
Comparable scores for this category are coming soon. One or both models do not have sourced results here yet.
GPT-4o mini
74.7
Ministral 3 3B (Reasoning)
59.7
Comparable scores for this category are coming soon. One or both models do not have sourced results here yet.
GPT-4o mini is ahead overall, 52 to 31. The biggest single separator in this matchup is HumanEval, where the scores are 87.2 and 16.
GPT-4o mini has the edge for knowledge tasks in this comparison, averaging 62 versus 25.2. Inside this category, MMLU is the benchmark that creates the most daylight between them.
GPT-4o mini has the edge for coding in this comparison, averaging 65 versus 7.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.
GPT-4o mini has the edge for reasoning in this comparison, averaging 49.4 versus 35.3. Inside this category, LongBench v2 is the benchmark that creates the most daylight between them.
GPT-4o mini has the edge for agentic tasks in this comparison, averaging 50.9 versus 34. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
GPT-4o mini has the edge for multimodal and grounded tasks in this comparison, averaging 60.2 versus 30.4. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
GPT-4o mini has the edge for multilingual tasks in this comparison, averaging 74.7 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.