Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Kimi K2.5 is clearly ahead on the aggregate, 39 to 23. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Kimi K2.5's sharpest advantage is in reasoning, where it averages 72.9 against 21.7. The single biggest benchmark swing on the page is SimpleQA, 21.7% to 74%.
Kimi K2.5 is also the more expensive model on tokens at $0.50 input / $2.80 output per 1M tokens, versus $0.30 input / $0.50 output per 1M tokens for Grok 3 Mini. That is roughly 5.6x on output cost alone.
Pick Kimi K2.5 if you want the stronger benchmark profile. Grok 3 Mini only becomes the better choice if you want the cheaper token bill.
Benchmark data for this category is coming soon.
Grok 3 Mini
41.5
Kimi K2.5
69
Benchmark data for this category is coming soon.
Grok 3 Mini
21.7
Kimi K2.5
72.9
Grok 3 Mini
74.1
Kimi K2.5
75
Benchmark data for this category is coming soon.
Benchmark data for this category is coming soon.
Grok 3 Mini
39.7
Kimi K2.5
76
Kimi K2.5 is ahead overall, 39 to 23. The biggest single separator in this matchup is SimpleQA, where the scores are 21.7% and 74%.
Kimi K2.5 has the edge for knowledge tasks in this comparison, averaging 75 versus 74.1. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Kimi K2.5 has the edge for coding in this comparison, averaging 69 versus 41.5. Grok 3 Mini stays close enough that the answer can still flip depending on your workload.
Kimi K2.5 has the edge for math in this comparison, averaging 76 versus 39.7. Inside this category, AIME 2024 is the benchmark that creates the most daylight between them.
Kimi K2.5 has the edge for reasoning in this comparison, averaging 72.9 versus 21.7. Inside this category, SimpleQA 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.