Enter opens the highlighted result. Esc closes.

MiniMax M2 vs MiniMax M2.5

MiniMax M2.5 leads MiniMax M2 on the Artificial Analysis Intelligence Index, 22.8 to 18.6, at about the same output price ($1.20 vs $1.02 per million). Updated October 22, 2026; 46 benchmarks have a result for both.

Replace MiniMax M2 Replace MiniMax M2.5
MiniMax M2MiniMax
Coding
No data
Overall
No data
Output $/M
$1.02
4
MiniMax M2.5MiniMax
Coding
37.2
Overall
37.3
Output $/M
$1.20
42
vs boards won of 46 shared

MiniMax M2 or MiniMax M2.5: which to pick

Only MiniMax M2.5 has a coding score, 37.2 on the Vibe Coding Index; MiniMax M2 has none, so the two cannot be ranked on coding.

Pick MiniMax M2 for output tokens 1.2x cheaper ($1.02 vs $1.20 per million) and work measured by Vending-Bench 2 or Kagi LLM Benchmark, where it leads.

Model AMiniMax M2MiniMaxNot benchmarked1 independent benchmark$0.702 per solved taskModel BMiniMax M2.5MiniMax37.2/1003 independent benchmarks$0.62 per solved task
Shared results
46
MiniMax M2 only
4
MiniMax M2.5 only
25
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
MiniMax M2 not benchmarked, MiniMax M2.5 37.2 from 3 benchmarks.
Widest category
Coding, 10.1 points apart on average over 7 shared benchmarks.
Widest single gap
LMArena WebDev: HTML: MiniMax M2 1306 against MiniMax M2.5 1414, 17.9 points apart on a 0 to 100 scale.
Output price
MiniMax M2 $1.02 against MiniMax M2.5 $1.20 per million output tokens, 1.2x apart.
Context window
Both take 205K tokens.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryMiniMax M2BarsMiniMax M2.5Ties
Coding07
Intelligence01
Agentic work11
Reasoning15
Math02
Knowledge13
Web design07
General116
All shared boards442
CodingLike-for-likeSix of these boards feed the Vibe Coding Index07
Coding benchmarks for MiniMax M2 and MiniMax M2.5
BenchmarkMiniMax M2MiniMax M2.5Leader
SWE-bench Verified61.0%75.8%MiniMax M2.5
SWE-bench MultilingualNot measured68.3%Only MiniMax M2.5 measured
LMArena WebDev12971386MiniMax M2.5
LiveCodeBench (Vals AI)Not measured79.2%Only MiniMax M2.5 measured
SWE-bench Verified (Vals AI)Not measured74.2%Only MiniMax M2.5 measured
Vibe Code Bench (Vals AI)Not measured14.8%Only MiniMax M2.5 measured
LiveCodeBench (Artificial Analysis run)82.6%Not measuredOnly MiniMax M2 measured
Terminal-Bench Hard (Artificial Analysis run)25.8%34.8%MiniMax M2.5
Terminal-Bench30.0%42.7%MiniMax M2.5
ALE-BenchNot measured618.2Only MiniMax M2.5 measured
LMArena Text: coding13711380MiniMax M2.5
LMArena Text: software and IT13731383MiniMax M2.5
LMArena WebDev: ReactNot measured1374Only MiniMax M2.5 measured
LMArena WebDev: HTML13061414MiniMax M2.5
SWE Atlas: Codebase QnANot measured10.3%Only MiniMax M2.5 measured
SWE Atlas: RefactoringNot measured19.5%Only MiniMax M2.5 measured
SWE Atlas: Test WritingNot measured18.6%Only MiniMax M2.5 measured
Neither model has a result on the other 26 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for MiniMax M2 and MiniMax M2.5
BenchmarkMiniMax M2MiniMax M2.5Leader
AA Intelligence Index18.622.8MiniMax M2.5
Epoch Capabilities IndexNot measured146.6Only MiniMax M2.5 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index11
Agentic work benchmarks for MiniMax M2 and MiniMax M2.5
BenchmarkMiniMax M2MiniMax M2.5Leader
τ²-Bench Telecom86.8%95.3%MiniMax M2.5
Vending-Bench 2$161-$23MiniMax M2
Neither model has a result on the other 40 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index15
Reasoning benchmarks for MiniMax M2 and MiniMax M2.5
BenchmarkMiniMax M2MiniMax M2.5Leader
GPQA DiamondNot measured82.1%Only MiniMax M2.5 measured
GPQA Diamond (Artificial Analysis run)77.7%84.8%MiniMax M2.5
Humanity's Last Exam (Artificial Analysis run)13.7%20.5%MiniMax M2.5
ARC-AGINot measured63.7%Only MiniMax M2.5 measured
ARC-AGI-2Not measured4.9%Only MiniMax M2.5 measured
LMArena Text: hard prompts13571372MiniMax M2.5
LMArena Text: expert13381378MiniMax M2.5
LMArena Text: hard prompts in English13761384MiniMax M2.5
Kagi LLM Benchmark57.8%55.2%MiniMax M2
Neither model has a result on the other 14 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for MiniMax M2 and MiniMax M2.5
BenchmarkMiniMax M2MiniMax M2.5Leader
AIMENot measured88.8%Only MiniMax M2.5 measured
AA Math Index78.3Not measuredOnly MiniMax M2 measured
AIME 2025 (Artificial Analysis run)78.3%Not measuredOnly MiniMax M2 measured
ProofBenchNot measured4.0%Only MiniMax M2.5 measured
LMArena Text: math13511378MiniMax M2.5
LMArena Text: mathematical work13681378MiniMax M2.5
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index13
Knowledge benchmarks for MiniMax M2 and MiniMax M2.5
BenchmarkMiniMax M2MiniMax M2.5Leader
MMLU ProNot measured80.1%Only MiniMax M2.5 measured
LegalBench (Vals AI)Not measured80.0%Only MiniMax M2.5 measured
CaseLaw (Vals AI)Not measured53.5%Only MiniMax M2.5 measured
CorpFin (Vals AI)Not measured59.6%Only MiniMax M2.5 measured
TaxEval (Vals AI)Not measured68.2%Only MiniMax M2.5 measured
MedQA (Vals AI)Not measured92.5%Only MiniMax M2.5 measured
MMLU Pro (Artificial Analysis run)82.0%Not measuredOnly MiniMax M2 measured
LMArena Text: science13481369MiniMax M2.5
LMArena Text: medicine and healthcare13671357MiniMax M2
LMArena Text: legal and government13491366MiniMax M2.5
LMArena Text: business and finance13361356MiniMax M2.5
Neither model has a result on the other 9 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index07
Web design benchmarks for MiniMax M2 and MiniMax M2.5
BenchmarkMiniMax M2MiniMax M2.5Leader
Design Arena: websites11431222MiniMax M2.5
Design Arena: UI components11341170MiniMax M2.5
Design Arena: game dev11181176MiniMax M2.5
Design Arena: data viz11401172MiniMax M2.5
Design Arena: 3D10981170MiniMax M2.5
Design Arena: all code categories11351207MiniMax M2.5
Design Arena: SVG11001150MiniMax M2.5
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index116
General benchmarks for MiniMax M2 and MiniMax M2.5
BenchmarkMiniMax M2MiniMax M2.5Leader
IFBench72.3%71.6%MiniMax M2
Long context reasoning (AA-LCR)64.3%73.3%MiniMax M2.5
CL-benchNot measured11.4%Only MiniMax M2.5 measured
CL-bench LifeNot measured6.3%Only MiniMax M2.5 measured
LMArena Text13401359MiniMax M2.5
LMArena Text: instruction following13291353MiniMax M2.5
LMArena Text: creative writing12861331MiniMax M2.5
LMArena Text: English13641380MiniMax M2.5
LMArena Text: non-English13131338MiniMax M2.5
LMArena Text: Chinese13661393MiniMax M2.5
LMArena Text: German13571361MiniMax M2.5
LMArena Text: JapaneseNot measured1171Only MiniMax M2.5 measured
LMArena Text: KoreanNot measured1233Only MiniMax M2.5 measured
LMArena Text: French13361363MiniMax M2.5
LMArena Text: Spanish13261353MiniMax M2.5
LMArena Text: Russian13301359MiniMax M2.5
LMArena Text: Polish13121334MiniMax M2.5
LMArena Text: multi-turn13611364MiniMax M2.5
LMArena Text: longer queries13311366MiniMax M2.5
LMArena Text: writing and language12951345MiniMax M2.5
LMArena Text: entertainment and media12981329MiniMax M2.5
Neither model has a result on the other 27 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for MiniMax M2 and MiniMax M2.5
BenchmarkMiniMax M2MiniMax M2.5Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

MiniMax M2$0.0008Fits in one request
MiniMax M2.5$0.0009Fits in one request

MiniMax M2 costs less

Repository review

50K input + 3K output tokens

MiniMax M2$0.016Fits in one request
MiniMax M2.5$0.019Fits in one request

MiniMax M2 costs less

Agent loop

200K input + 10K output tokens

MiniMax M2$0.061Exceeds the context window
MiniMax M2.5$0.072Exceeds the context window

MiniMax M2 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of MiniMax M2 and MiniMax M2.5
SpecificationMiniMax M2MiniMax M2.5
MakerMiniMaxMiniMax
WeightsOpen weightsOpen weights
ReleasedOctober 23, 2025February 12, 2026
Context window205K205K
Input $/Mtok$0.255$0.30
Output $/Mtok$1.02$1.20
Coding indexNot benchmarked37.2
Independent measurements behind it13
Cost per solved task$0.702$0.62
SourceOpenRouter listingOpenRouter listing

Questions

Is MiniMax M2 better than MiniMax M2.5 for coding?

Only MiniMax M2.5 has a coding score, 37.2 on the Vibe Coding Index; MiniMax M2 has none, so the two cannot be ranked on coding. Pick MiniMax M2 for output tokens 1.2x cheaper ($1.02 vs $1.20 per million) and work measured by Vending-Bench 2 or Kagi LLM Benchmark, where it leads.

Which is cheaper, MiniMax M2 or MiniMax M2.5?

MiniMax M2 costs $0.255 per million input tokens and $1.02 per million output tokens; MiniMax M2.5 costs $0.30 and $1.20. Per solved SWE-bench Verified task, MiniMax M2 costs $0.702 and MiniMax M2.5 $0.62.

Which has the bigger context window, MiniMax M2 or MiniMax M2.5?

Neither: both take 205K tokens.

How many benchmarks is the MiniMax M2 vs MiniMax M2.5 comparison based on?

46 benchmarks have a result for both models as of October 22, 2026. MiniMax M2's coding index rests on 1 independent benchmark and MiniMax M2.5's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next