Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
MiniMax M2.5 is clearly ahead on the aggregate, 59 to 38. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
MiniMax M2.5's sharpest advantage is in mathematics, where it averages 76.1 against 50.4. The single biggest benchmark swing on the page is SWE-bench Verified, 45 to 18.
MiniMax M2.5 is also the more expensive model on tokens at $0.30 input / $1.20 output per 1M tokens, versus $0.03 input / $0.12 output per 1M tokens for LFM2-24B-A2B. That is roughly 10.0x on output cost alone. MiniMax M2.5 gives you the larger context window at 128K, compared with 32K for LFM2-24B-A2B.
Pick MiniMax M2.5 if you want the stronger benchmark profile. LFM2-24B-A2B only becomes the better choice if you want the cheaper token bill.
MiniMax M2.5
53.4
LFM2-24B-A2B
33.4
MiniMax M2.5
38.7
LFM2-24B-A2B
18
MiniMax M2.5
62
LFM2-24B-A2B
41.7
MiniMax M2.5
69.6
LFM2-24B-A2B
46.6
MiniMax M2.5
55.3
LFM2-24B-A2B
35.6
MiniMax M2.5
85
LFM2-24B-A2B
68
MiniMax M2.5
82.1
LFM2-24B-A2B
61.4
MiniMax M2.5
76.1
LFM2-24B-A2B
50.4
MiniMax M2.5 is ahead overall, 59 to 38. The biggest single separator in this matchup is SWE-bench Verified, where the scores are 45 and 18.
MiniMax M2.5 has the edge for knowledge tasks in this comparison, averaging 55.3 versus 35.6. Inside this category, MMLU is the benchmark that creates the most daylight between them.
MiniMax M2.5 has the edge for coding in this comparison, averaging 38.7 versus 18. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.
MiniMax M2.5 has the edge for math in this comparison, averaging 76.1 versus 50.4. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.
MiniMax M2.5 has the edge for reasoning in this comparison, averaging 69.6 versus 46.6. Inside this category, SimpleQA is the benchmark that creates the most daylight between them.
MiniMax M2.5 has the edge for agentic tasks in this comparison, averaging 53.4 versus 33.4. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.
MiniMax M2.5 has the edge for multimodal and grounded tasks in this comparison, averaging 62 versus 41.7. Inside this category, OfficeQA Pro is the benchmark that creates the most daylight between them.
MiniMax M2.5 has the edge for instruction following in this comparison, averaging 85 versus 68. Inside this category, IFEval is the benchmark that creates the most daylight between them.
MiniMax M2.5 has the edge for multilingual tasks in this comparison, averaging 82.1 versus 61.4. Inside this category, MMLU-ProX 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.