Enter opens the highlighted result. Esc closes.

MiniMax M3 vs Mistral Large 4

Mistral Large 4 leads MiniMax M3 on coding, 56.4 to 39.2 on the Vibe Coding Index, and MiniMax M3 costs 3.5x less per output token ($1.20 vs $4.18 per million). Updated October 22, 2026; 73 benchmarks have a result for both.

Replace MiniMax M3 Replace Mistral Large 4
MiniMax M3MiniMax
Coding
39.2
Overall
51.7
Output $/M
$1.20
27
Mistral Large 4Mistral AI
Coding
56.4
Overall
54.6
Output $/M
$4.18
44
vs boards won of 73 shared, 2 tied

MiniMax M3 or Mistral Large 4: which to pick

Pick Mistral Large 4 for coding: it scores 56.4 on the Vibe Coding Index against 39.2 for MiniMax M3, though on 5 benchmarks against 6.

Pick MiniMax M3 for output tokens 3.5x cheaper ($1.20 vs $4.18 per million) and work measured by LiveBench coding or SWE-bench Multimodal (Vals AI), where it leads.

Model AMiniMax M3MiniMax39.2/1006 independent benchmarks$0.56 per solved taskModel BMistral Large 4Mistral AI56.4/1005 independent benchmarksNo cost per solved task
Shared results
73
MiniMax M3 only
43
Mistral Large 4 only
8
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 M3 39.2 from 6 benchmarks, Mistral Large 4 56.4 from 5 benchmarks.
Widest category
Coding, 13.3 points apart on average over 14 shared benchmarks.
Widest single gap
LiveBench research: MiniMax M3 68.0% against Mistral Large 4 28.1%, 39.9 points apart on a 0 to 100 scale.
Output price
MiniMax M3 $1.20 against Mistral Large 4 $4.18 per million output tokens, 3.5x apart.
Context window
Both take 1.05M 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 M3BarsMistral Large 4Ties
Coding212
Intelligence02
Agentic work241 tie
Reasoning54
Math13
Knowledge41
Web design171 tie
General1211
All shared boards27442 ties
CodingLike-for-likeSix of these boards feed the Vibe Coding Index212
Coding benchmarks for MiniMax M3 and Mistral Large 4
BenchmarkMiniMax M3Mistral Large 4Leader
LiveBench coding70.9%60.0%MiniMax M3
LMArena WebDev14821543Mistral Large 4
AA Coding Index58.6Not measuredOnly MiniMax M3 measured
LiveCodeBench (Vals AI)82.2%Not measuredOnly MiniMax M3 measured
SWE-bench Verified (Vals AI)75.0%Not measuredOnly MiniMax M3 measured
Terminal-Bench 4 (Vals AI)1.0%27.8%Mistral Large 4
Vibe Code Bench (Vals AI)47.6%78.4%Mistral Large 4
Code migration (Vals AI)19.9%30.6%Mistral Large 4
IOI (Vals AI)Not measured45.3%Only Mistral Large 4 measured
Terminal-Bench 2.1 (Vals AI)53.6%Not measuredOnly MiniMax M3 measured
SWE-bench Multimodal (Vals AI)33.3%33.1%MiniMax M3
Vibe Code Bench 1-100 (Vals AI)9.2%14.3%Mistral Large 4
SkillsBench (Vals AI)51.5%Not measuredOnly MiniMax M3 measured
CyberBench (Vals AI)65.0%Not measuredOnly MiniMax M3 measured
SciCode47.1%54.2%Mistral Large 4
Terminal-Bench Hard (Artificial Analysis run)42.4%Not measuredOnly MiniMax M3 measured
FrontierCode14.7%39.4%Mistral Large 4
ALE-Bench640.01213.2Mistral Large 4
LMArena Text: coding14701478Mistral Large 4
LMArena Text: software and IT14661467Mistral Large 4
LMArena WebDev: React14781559Mistral Large 4
LMArena WebDev: HTML14551487Mistral Large 4
LMArena WebDev: from an image1470Not measuredOnly MiniMax M3 measured
Neither model has a result on the other 16 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for MiniMax M3 and Mistral Large 4
BenchmarkMiniMax M3Mistral Large 4Leader
AA Intelligence Index29.238.4Mistral Large 4
Epoch Capabilities Index146.9150.0Mistral Large 4
Agentic workLike-for-likePublic boards, shown as published, kept out of the index24
Agentic work benchmarks for MiniMax M3 and Mistral Large 4
BenchmarkMiniMax M3Mistral Large 4Leader
AA Agentic Index29.5Not measuredOnly MiniMax M3 measured
Design Arena: agents building web apps1200Not measuredOnly MiniMax M3 measured
Design Arena: agents building full-stack apps1197Not measuredOnly MiniMax M3 measured
Design Arena: agents building mobile apps1169Not measuredOnly MiniMax M3 measured
Design Arena: agents building Android apps1159Not measuredOnly MiniMax M3 measured
Design Arena: agents building games1158Not measuredOnly MiniMax M3 measured
Design Arena: HTML slides1130Not measuredOnly MiniMax M3 measured
Design Arena: PowerPoint slides1181Not measuredOnly MiniMax M3 measured
Design Arena: spreadsheets1028Not measuredOnly MiniMax M3 measured
LiveBench research68.0%28.1%MiniMax M3
LiveBench browser8.0%8.0%Tie
LiveBench long agentic12.4%11.8%MiniMax M3
Finance Agent (Vals AI)48.3%54.7%Mistral Large 4
Tax Agent Bench (Vals AI)20.6%30.4%Mistral Large 4
Legal Research Bench (Vals AI)29.8%31.7%Mistral Large 4
Harvey's Legal Agent Benchmark (Vals AI)Not measured17.5%Only Mistral Large 4 measured
Excel Modeling Benchmark (Vals AI)47.8%56.2%Mistral Large 4
BioMysteryBench (Vals AI)Not measured67.0%Only Mistral Large 4 measured
τ²-Bench Telecom88.9%Not measuredOnly MiniMax M3 measured
APEX-Agents37.7%Not measuredOnly MiniMax M3 measured
OSWorld 2.04.6%Not measuredOnly MiniMax M3 measured
Vending-Bench 2$2,158Not measuredOnly MiniMax M3 measured
Neither model has a result on the other 19 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index54
Reasoning benchmarks for MiniMax M3 and Mistral Large 4
BenchmarkMiniMax M3Mistral Large 4Leader
GPQA Diamond92.7%Not measuredOnly MiniMax M3 measured
LiveBench reasoning74.5%83.9%Mistral Large 4
MysteryMechanism (Vals AI)Not measured12.2%Only Mistral Large 4 measured
MMMU Pro (Vals AI)81.2%Not measuredOnly MiniMax M3 measured
GPQA Diamond (Artificial Analysis run)92.9%Not measuredOnly MiniMax M3 measured
Humanity's Last Exam (Artificial Analysis run)39.0%35.0%MiniMax M3
GPQA Diamond (Epoch AI run)90.9%88.1%MiniMax M3
SimpleBench45.8%Not measuredOnly MiniMax M3 measured
Chess puzzles13.0%33.0%Mistral Large 4
Mystery game puzzles8.0%17.0%Mistral Large 4
LMArena Text: hard prompts14471444MiniMax M3
LMArena Text: expert14611444MiniMax M3
LMArena Text: hard prompts in English14551442MiniMax M3
LMArena Vision: diagrams12661288Mistral Large 4
LMArena Vision: homework1268Not measuredOnly MiniMax M3 measured
Neither model has a result on the other 7 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index13
Math benchmarks for MiniMax M3 and Mistral Large 4
BenchmarkMiniMax M3Mistral Large 4Leader
OTIS Mock AIME71.1%98.3%Mistral Large 4
FrontierMath, tiers 1 to 3Not measured58.6%Only Mistral Large 4 measured
FrontierMath, tier 4Not measured26.8%Only Mistral Large 4 measured
ProofBench18.0%10.0%MiniMax M3
LMArena Text: math14291469Mistral Large 4
LMArena Text: mathematical work14481462Mistral Large 4
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index41
Knowledge benchmarks for MiniMax M3 and Mistral Large 4
BenchmarkMiniMax M3Mistral Large 4Leader
MMLU Pro84.2%Not measuredOnly MiniMax M3 measured
LegalBench (Vals AI)85.4%Not measuredOnly MiniMax M3 measured
CorpFin (Vals AI)68.1%Not measuredOnly MiniMax M3 measured
TaxEval (Vals AI)72.7%Not measuredOnly MiniMax M3 measured
MortgageTax (Vals AI)68.4%Not measuredOnly MiniMax M3 measured
MedCode (Vals AI)46.3%40.7%MiniMax M3
SimpleQA VerifiedNot measured20.0%Only Mistral Large 4 measured
LMArena Text: science14521448MiniMax M3
LMArena Text: medicine and healthcare14411433MiniMax M3
LMArena Text: legal and government14371425MiniMax M3
LMArena Text: business and finance14341439Mistral Large 4
Neither model has a result on the other 7 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index17
Web design benchmarks for MiniMax M3 and Mistral Large 4
BenchmarkMiniMax M3Mistral Large 4Leader
Design Arena: websites12571257Tie
Design Arena: UI components12421272Mistral Large 4
Design Arena: game dev12101225Mistral Large 4
Design Arena: data viz12361275Mistral Large 4
Design Arena: 3D12081252Mistral Large 4
Design Arena: all code categories12471259Mistral Large 4
Design Arena: SVG11611232Mistral Large 4
Design Arena: ASCII art11611202Mistral Large 4
LiveBench design66.2%65.6%MiniMax M3
GeneralLike-for-likePublic boards, shown as published, kept out of the index1211
General benchmarks for MiniMax M3 and Mistral Large 4
BenchmarkMiniMax M3Mistral Large 4Leader
LiveBench data analysis76.2%76.5%Mistral Large 4
Public Benefits Bench (Vals AI)64.1%Not measuredOnly MiniMax M3 measured
SAGE (Vals AI)50.6%Not measuredOnly MiniMax M3 measured
MedScribe (Vals AI)87.3%80.4%MiniMax M3
Vals Multimodal Index60.0%Not measuredOnly MiniMax M3 measured
IFBench82.9%Not measuredOnly MiniMax M3 measured
Long context reasoning (AA-LCR)83.0%81.3%MiniMax M3
DTBench78.9%88.2%Mistral Large 4
LMCA33.7%35.8%Mistral Large 4
Surface Evolver Bench55.0%28.1%MiniMax M3
Furniture assemblyNot measured20.8%Only Mistral Large 4 measured
LMArena Text14331431MiniMax M3
LMArena Text: instruction following14331422MiniMax M3
LMArena Text: creative writing14041363MiniMax M3
LMArena Vision12511272Mistral Large 4
LMArena Text: English14431433MiniMax M3
LMArena Text: non-English14201422Mistral Large 4
LMArena Text: Chinese14661492Mistral Large 4
LMArena Text: German1428Not measuredOnly MiniMax M3 measured
LMArena Text: Japanese13801415Mistral Large 4
LMArena Text: Korean1373Not measuredOnly MiniMax M3 measured
LMArena Text: French14441492Mistral Large 4
LMArena Text: Spanish14331455Mistral Large 4
LMArena Text: Russian14261417MiniMax M3
LMArena Text: Polish1423Not measuredOnly MiniMax M3 measured
LMArena Text: multi-turn14431435MiniMax M3
LMArena Text: longer queries14461428MiniMax M3
LMArena Text: writing and language14191386MiniMax M3
LMArena Text: entertainment and media14031378MiniMax M3
LMArena Vision: English12631283Mistral Large 4
LMArena Vision: Chinese1311Not measuredOnly MiniMax M3 measured
LMArena Vision: reading text12591284Mistral Large 4
LMArena Vision: humor1235Not measuredOnly MiniMax M3 measured
LMArena Vision: recognizing things1197Not measuredOnly MiniMax M3 measured
LMArena Vision: creative writing1234Not measuredOnly MiniMax M3 measured
LMArena Document1435Not measuredOnly MiniMax M3 measured
Neither model has a result on the other 10 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for MiniMax M3 and Mistral Large 4
BenchmarkMiniMax M3Mistral Large 4Leader
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 M3$0.0009Fits in one request
Mistral Large 4$0.0034Fits in one request

MiniMax M3 costs less

Repository review

50K input + 3K output tokens

MiniMax M3$0.019Fits in one request
Mistral Large 4$0.081Fits in one request

MiniMax M3 costs less

Agent loop

200K input + 10K output tokens

MiniMax M3$0.072Fits in one request
Mistral Large 4$0.314Fits in one request

MiniMax M3 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 M3 and Mistral Large 4
SpecificationMiniMax M3Mistral Large 4
MakerMiniMaxMistral AI
WeightsOpen weightsNo public weights listed
ReleasedMay 31, 2026October 6, 2026
Context window1.05M1.05M
Input $/Mtok$0.30$1.36
Output $/Mtok$1.20$4.18
Coding index39.256.4
Independent measurements behind it65
Cost per solved task$0.56Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is MiniMax M3 better than Mistral Large 4 for coding?

Pick Mistral Large 4 for coding: it scores 56.4 on the Vibe Coding Index against 39.2 for MiniMax M3, though on 5 benchmarks against 6. Pick MiniMax M3 for output tokens 3.5x cheaper ($1.20 vs $4.18 per million) and work measured by LiveBench coding or SWE-bench Multimodal (Vals AI), where it leads.

Which is cheaper, MiniMax M3 or Mistral Large 4?

MiniMax M3 costs $0.30 per million input tokens and $1.20 per million output tokens; Mistral Large 4 costs $1.36 and $4.18. Cost per solved task is known for MiniMax M3 only.

Which has the bigger context window, MiniMax M3 or Mistral Large 4?

Neither: both take 1.05M tokens.

How many benchmarks is the MiniMax M3 vs Mistral Large 4 comparison based on?

73 benchmarks have a result for both models as of October 22, 2026. MiniMax M3's coding index rests on 6 independent benchmarks and Mistral Large 4's on 5. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/minimax-minimax-m3-vs-mistral-large-4-0" title="MiniMax M3 vs Mistral Large 4, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/minimax-minimax-m3-vs-mistral-large-4-0">LLMBenchmark.ai, MiniMax M3 vs Mistral Large 4</a></p>