Enter opens the highlighted result. Esc closes.

MiniMax M2.5 vs Qwen2.5 72B Instruct

MiniMax M2.5 leads Qwen2.5 72B Instruct on the Artificial Analysis Intelligence Index, 22.8 to 7.7, and Qwen2.5 72B Instruct costs 3.0x less per output token ($0.40 vs $1.20 per million). Updated October 22, 2026; 34 benchmarks have a result for both.

Replace MiniMax M2.5 Replace Qwen2.5 72B Instruct
MiniMax M2.5MiniMax
Coding
37.2
Overall
37.3
Output $/M
$1.20
33
Qwen2.5 72B InstructQwen
Coding
No data
Overall
26.5
Output $/M
$0.40
1
vs boards won of 34 shared

MiniMax M2.5 or Qwen2.5 72B Instruct: which to pick

Only MiniMax M2.5 has a coding score, 37.2 on the Vibe Coding Index; Qwen2.5 72B Instruct has none, so the two cannot be ranked on coding.

Pick Qwen2.5 72B Instruct for output tokens 3.0x cheaper ($0.40 vs $1.20 per million) and work measured by LMArena Text: Japanese, where it leads.

Model AMiniMax M2.5MiniMax37.2/1003 independent benchmarks$0.62 per solved taskModel BQwen2.5 72B InstructQwenNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
34
MiniMax M2.5 only
37
Qwen2.5 72B Instruct only
19
Like-for-like categories
7 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 M2.5 37.2 from 3 benchmarks, Qwen2.5 72B Instruct not benchmarked.
Widest category
Agentic work, 60.8 points apart on average over 1 shared benchmark.
Widest single gap
τ²-Bench Telecom: MiniMax M2.5 95.3% against Qwen2.5 72B Instruct 34.5%, 60.8 points apart on a 0 to 100 scale.
Output price
MiniMax M2.5 $1.20 against Qwen2.5 72B Instruct $0.40 per million output tokens, 3.0x apart.
Context window
MiniMax M2.5 takes 205K tokens, Qwen2.5 72B Instruct 33K.

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 M2.5BarsQwen2.5 72B InstructTies
Coding30
Intelligence20
Agentic work10
Reasoning50
Math20
Knowledge40
General161
All shared boards331
CodingLike-for-likeSix of these boards feed the Vibe Coding Index30
Coding benchmarks for MiniMax M2.5 and Qwen2.5 72B Instruct
BenchmarkMiniMax M2.5Qwen2.5 72B InstructLeader
SWE-bench Verified75.8%Not measuredOnly MiniMax M2.5 measured
SWE-bench Multilingual68.3%Not measuredOnly MiniMax M2.5 measured
LMArena WebDev1386Not measuredOnly MiniMax M2.5 measured
LiveCodeBench (Vals AI)79.2%Not measuredOnly MiniMax M2.5 measured
SWE-bench Verified (Vals AI)74.2%Not measuredOnly MiniMax M2.5 measured
Vibe Code Bench (Vals AI)14.8%Not measuredOnly MiniMax M2.5 measured
LiveCodeBench (Artificial Analysis run)Not measured27.6%Only Qwen2.5 72B Instruct measured
Terminal-Bench Hard (Artificial Analysis run)34.8%4.5%MiniMax M2.5
Terminal-Bench42.7%Not measuredOnly MiniMax M2.5 measured
WeirdMLNot measured16.0%Only Qwen2.5 72B Instruct measured
ALE-Bench618.2Not measuredOnly MiniMax M2.5 measured
LMArena Text: coding13801292MiniMax M2.5
LMArena Text: software and IT13831292MiniMax M2.5
LMArena WebDev: React1374Not measuredOnly MiniMax M2.5 measured
LMArena WebDev: HTML1414Not measuredOnly MiniMax M2.5 measured
SWE Atlas: Codebase QnA10.3%Not measuredOnly MiniMax M2.5 measured
SWE Atlas: Refactoring19.5%Not measuredOnly MiniMax M2.5 measured
SWE Atlas: Test Writing18.6%Not measuredOnly MiniMax M2.5 measured
Neither model has a result on the other 25 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for MiniMax M2.5 and Qwen2.5 72B Instruct
BenchmarkMiniMax M2.5Qwen2.5 72B InstructLeader
AA Intelligence Index22.87.7MiniMax M2.5
Epoch Capabilities Index146.6129.0MiniMax M2.5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for MiniMax M2.5 and Qwen2.5 72B Instruct
BenchmarkMiniMax M2.5Qwen2.5 72B InstructLeader
τ²-Bench Telecom95.3%34.5%MiniMax M2.5
METR time horizonsNot measured35.8%Only Qwen2.5 72B Instruct measured
BALROGNot measured16.2%Only Qwen2.5 72B Instruct measured
TheAgentCompanyNot measured5.7%Only Qwen2.5 72B Instruct measured
Vending-Bench 2-$23Not measuredOnly MiniMax M2.5 measured
Neither model has a result on the other 37 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index50
Reasoning benchmarks for MiniMax M2.5 and Qwen2.5 72B Instruct
BenchmarkMiniMax M2.5Qwen2.5 72B InstructLeader
GPQA Diamond82.1%Not measuredOnly MiniMax M2.5 measured
GPQA Diamond (Artificial Analysis run)84.8%49.1%MiniMax M2.5
Humanity's Last Exam (Artificial Analysis run)20.5%3.6%MiniMax M2.5
GPQA Diamond (Epoch AI run)Not measured49.1%Only Qwen2.5 72B Instruct measured
ARC-AGI63.7%Not measuredOnly MiniMax M2.5 measured
ARC-AGI-24.9%Not measuredOnly MiniMax M2.5 measured
BIG-Bench HardNot measured79.8%Only Qwen2.5 72B Instruct measured
LMArena Text: hard prompts13721271MiniMax M2.5
LMArena Text: expert13781245MiniMax M2.5
LMArena Text: hard prompts in English13841279MiniMax M2.5
Kagi LLM Benchmark55.2%Not measuredOnly MiniMax M2.5 measured
Neither model has a result on the other 12 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index20
Math benchmarks for MiniMax M2.5 and Qwen2.5 72B Instruct
BenchmarkMiniMax M2.5Qwen2.5 72B InstructLeader
AIME88.8%Not measuredOnly MiniMax M2.5 measured
AA Math IndexNot measured14.0Only Qwen2.5 72B Instruct measured
AIME 2025 (Artificial Analysis run)Not measured14.0%Only Qwen2.5 72B Instruct measured
OTIS Mock AIMENot measured8.1%Only Qwen2.5 72B Instruct measured
ProofBench4.0%Not measuredOnly MiniMax M2.5 measured
MATH level 5Not measured63.2%Only Qwen2.5 72B Instruct measured
LMArena Text: math13781283MiniMax M2.5
LMArena Text: mathematical work13781290MiniMax M2.5
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
Knowledge benchmarks for MiniMax M2.5 and Qwen2.5 72B Instruct
BenchmarkMiniMax M2.5Qwen2.5 72B InstructLeader
MMLU Pro80.1%Not measuredOnly MiniMax M2.5 measured
LegalBench (Vals AI)80.0%Not measuredOnly MiniMax M2.5 measured
CaseLaw (Vals AI)53.5%Not measuredOnly MiniMax M2.5 measured
CorpFin (Vals AI)59.6%Not measuredOnly MiniMax M2.5 measured
TaxEval (Vals AI)68.2%Not measuredOnly MiniMax M2.5 measured
MedQA (Vals AI)92.5%Not measuredOnly MiniMax M2.5 measured
MMLU Pro (Artificial Analysis run)Not measured72.0%Only Qwen2.5 72B Instruct measured
MMLUNot measured85.3%Only Qwen2.5 72B Instruct measured
AI2 Reasoning ChallengeNot measured94.5%Only Qwen2.5 72B Instruct measured
LMArena Text: science13691275MiniMax M2.5
LMArena Text: medicine and healthcare13571258MiniMax M2.5
LMArena Text: legal and government13661288MiniMax M2.5
LMArena Text: business and finance13561253MiniMax M2.5
Neither model has a result on the other 7 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Web design benchmarks for MiniMax M2.5 and Qwen2.5 72B Instruct
BenchmarkMiniMax M2.5Qwen2.5 72B InstructLeader
Design Arena: websites1222Not measuredOnly MiniMax M2.5 measured
Design Arena: UI components1170Not measuredOnly MiniMax M2.5 measured
Design Arena: game dev1176Not measuredOnly MiniMax M2.5 measured
Design Arena: data viz1172Not measuredOnly MiniMax M2.5 measured
Design Arena: 3D1170Not measuredOnly MiniMax M2.5 measured
Design Arena: all code categories1207Not measuredOnly MiniMax M2.5 measured
Design Arena: SVG1150Not measuredOnly MiniMax M2.5 measured
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index161
General benchmarks for MiniMax M2.5 and Qwen2.5 72B Instruct
BenchmarkMiniMax M2.5Qwen2.5 72B InstructLeader
IFBench71.6%36.9%MiniMax M2.5
Long context reasoning (AA-LCR)73.3%Not measuredOnly MiniMax M2.5 measured
DTBenchNot measured62.9%Only Qwen2.5 72B Instruct measured
LMCANot measured13.4%Only Qwen2.5 72B Instruct measured
CL-bench11.4%Not measuredOnly MiniMax M2.5 measured
HellaSwagNot measured84.8%Only Qwen2.5 72B Instruct measured
WinoGrandeNot measured82.3%Only Qwen2.5 72B Instruct measured
PIQANot measured82.6%Only Qwen2.5 72B Instruct measured
CL-bench Life6.3%Not measuredOnly MiniMax M2.5 measured
LMArena Text13591269MiniMax M2.5
LMArena Text: instruction following13531254MiniMax M2.5
LMArena Text: creative writing13311221MiniMax M2.5
LMArena Text: English13801283MiniMax M2.5
LMArena Text: non-English13381252MiniMax M2.5
LMArena Text: Chinese13931273MiniMax M2.5
LMArena Text: German13611234MiniMax M2.5
LMArena Text: Japanese11711180Qwen2.5 72B Instruct
LMArena Text: Korean12331188MiniMax M2.5
LMArena Text: French13631280MiniMax M2.5
LMArena Text: Spanish13531255MiniMax M2.5
LMArena Text: Russian13591264MiniMax M2.5
LMArena Text: Polish1334Not measuredOnly MiniMax M2.5 measured
LMArena Text: multi-turn13641272MiniMax M2.5
LMArena Text: longer queries13661281MiniMax M2.5
LMArena Text: writing and language13451247MiniMax M2.5
LMArena Text: entertainment and media13291211MiniMax M2.5
Neither model has a result on the other 22 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for MiniMax M2.5 and Qwen2.5 72B Instruct
BenchmarkMiniMax M2.5Qwen2.5 72B InstructLeader
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 M2.5$0.0009Fits in one request
Qwen2.5 72B Instruct$0.0006Fits in one request

Qwen2.5 72B Instruct costs less

Repository review

50K input + 3K output tokens

MiniMax M2.5$0.019Fits in one request
Qwen2.5 72B Instruct$0.019Exceeds the context window

MiniMax M2.5 costs less

Agent loop

200K input + 10K output tokens

MiniMax M2.5$0.072Exceeds the context window
Qwen2.5 72B Instruct$0.076Exceeds the context window

MiniMax M2.5 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 M2.5 and Qwen2.5 72B Instruct
SpecificationMiniMax M2.5Qwen2.5 72B Instruct
MakerMiniMaxQwen
WeightsOpen weightsOpen weights
ReleasedFebruary 12, 2026September 19, 2024
Context window205K33K
Input $/Mtok$0.30$0.36
Output $/Mtok$1.20$0.40
Coding index37.2Not benchmarked
Independent measurements behind it30
Cost per solved task$0.62Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is MiniMax M2.5 better than Qwen2.5 72B Instruct for coding?

Only MiniMax M2.5 has a coding score, 37.2 on the Vibe Coding Index; Qwen2.5 72B Instruct has none, so the two cannot be ranked on coding. Pick Qwen2.5 72B Instruct for output tokens 3.0x cheaper ($0.40 vs $1.20 per million) and work measured by LMArena Text: Japanese, where it leads.

Which is cheaper, MiniMax M2.5 or Qwen2.5 72B Instruct?

MiniMax M2.5 costs $0.30 per million input tokens and $1.20 per million output tokens; Qwen2.5 72B Instruct costs $0.36 and $0.40. Cost per solved task is known for MiniMax M2.5 only.

Which has the bigger context window, MiniMax M2.5 or Qwen2.5 72B Instruct?

MiniMax M2.5, with 205K tokens against 33K.

How many benchmarks is the MiniMax M2.5 vs Qwen2.5 72B Instruct comparison based on?

34 benchmarks have a result for both models as of October 22, 2026. MiniMax M2.5's coding index rests on 3 independent benchmarks and Qwen2.5 72B Instruct's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next