Kimi K2.5 vs MiniMax M2.1
Wins 5 of 7 areas
Coding · Reasoning · Facts · Math · Long documents
Wins 0 of 7 areas
—
Kimi K2.5 is the stronger all-rounder.MiniMax M2.1 is cheaper.
Scores updated · 37 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- CodingWriting and fixing software50Kimi K2.55 of 6 tests · 1 tie
- ReasoningHard problems that need careful thinking30Kimi K2.53 of 3 tests
- FactsGetting facts right instead of making them up21Kimi K2.52 of 3 tests
- MathCompetition and research-level math10Kimi K2.51 of 1 test
- Long documentsFinding answers in very long texts10Kimi K2.51 of 1 test
- AgentsCarrying out multi-step tasks on its own11Even1 each
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
Math, long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
MiniMax M2.1 costs 58% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Kimi K2.5 pulls ahead
- Olympiad math problems with short answersIMOAnswerBench+21.4points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+14points ahead
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+13.9points ahead
Where MiniMax M2.1 pulls ahead
- Real work tasks from 44 professionsGDPVal+12.5points ahead
Every test, side by side
All 37 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingKimi K2.5
- SWE-bench ProKimi K2.5 by 13.950.736.8+13.9
- SWE-bench VerifiedKimi K2.5 by 9.876.867+9.8
- SciCodeKimi K2.5 by 8.34940.7+8.3
- Terminal-Bench HardKimi K2.5 by 634.828.8+6
- LMArena · WebDevKimi K2.5 by 52 rating points14361384+52 rating
- SWE-bench Multilingualtie7372.5tie
AgentsEven
- BrowseCompKimi K2.5 by 12.974.962+12.9
- GDPValMiniMax M2.1 by 12.517.229.7+12.5
ReasoningKimi K2.5
- Humanity's Last ExamKimi K2.5 by 7.530.723.2+7.5
- GPQA DiamondKimi K2.5 by 4.987.983+4.9
- CritPtKimi K2.5 by 2.83.10.3+2.8
FactsKimi K2.5
- AA-Omniscience · AccuracyKimi K2.5 by 1435.221.2+14
- AA-Omniscience · Non-hallucinationKimi K2.5 by 2.834.331.5+2.8
- Vectara HHEM hallucination ratelower is betterMiniMax M2.1 by 2.414.211.8+2.4
Other results20 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- xbench-DeepSearchKimi K2.5 by 33.776.743+33.7
- AA Agentic IndexMiniMax M2.1 by 25.721.747.4+25.7
- AA-OmniscienceKimi K2.5 by 25.6-7.3-32.9+25.6
- HMMT Feb. 2025Kimi K2.5 by 24.495.471+24.4
- ToolathlonMiniMax M2.1 by 15.727.843.5+15.7
- AIME 2025Kimi K2.5 by 15.196.181+15.1
- HMMT Nov. 2025Kimi K2.5 by 14.989.274.3+14.9
- BrowseComp-ZHKimi K2.5 by 14.562.347.8+14.5
- Artificial Analysis Coding IndexKimi K2.5 by 1446.832.8+14
- browsecomp_with_context_managerKimi K2.5 by 12.974.962+12.9
- GAIA (no file)Kimi K2.5 by 11.675.964.3+11.6
- τ²-Bench Telecom (AA run)Kimi K2.5 by 10.595.985.4+10.5
- vectara_answer_rateMiniMax M2.1 by 6.392.298.5+6.3
- Terminal-Bench 2.0Kimi K2.5 by 2.950.847.9+2.9
- AA IntelligenceKimi K2.5 by 2.623.520.9+2.6
- vectara_factual_consistencyMiniMax M2.1 by 2.485.888.2+2.4
- τ²-BenchMiniMax M2.1 by 1.685.487+1.6
- MMLU-Protie87.188tie
- ResearchRubricstie59.560.2tie
- vectara_avg_summary_lengthtie112106.9tie
Questions people ask
Which is better, Kimi K2.5 or MiniMax M2.1?
Kimi K2.5 wins five of the seven areas where both have results: coding, reasoning, facts, math and long documents. MiniMax M2.1 wins none, but costs 58% less. They are level on agents and following instructions.
Which is better for coding?
Kimi K2.5. It wins 5 of the 6 coding tests both models report; MiniMax M2.1 wins none, and 1 is a tie.
Which is cheaper?
Kimi K2.5 costs $0.60 per million input tokens and $3.00 per million output tokens; MiniMax M2.1 costs $0.30 and $1.20. That makes MiniMax M2.1 about 58% cheaper for the same work.
How do you compare the two?
We use the 37 benchmark tests both models have published scores on. The verdict counts the 17 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.