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 23, 2026; 74 benchmarks have a result for both.
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.
- Shared results
- 74
- MiniMax M3 only
- 43
- Mistral Large 4 only
- 9
- 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.
Coding
Vibe Coding Index, six public coding benchmarks
Mistral Large 4
Mistral Large 4 scores 56.4 against 39.2.
6 and 5 benchmarks behind the scores
Intelligence
AA Intelligence Index
Mistral Large 4
Mistral Large 4 leads, 38.4 against 29.2.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only MiniMax M3 has a AA Agentic Index result, 29.5.
Directional only
Reasoning
GPQA Diamond
No call
Only MiniMax M3 has a GPQA Diamond result, 92.7%.
Directional only
Knowledge
MMLU Pro
No call
Only MiniMax M3 has a MMLU Pro result, 84.2%.
Directional only
Web design
Design Arena: websites
Mistral Large 4
Mistral Large 4 leads, 1258 against 1257.
Same board, both measured
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
MiniMax M3
MiniMax M3 $0.0009, Mistral Large 4 $0.0034 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only MiniMax M3 has the number, $0.56.
Directional only
At a glance
- Coding index
- MiniMax M3 39.2 from 6 benchmarks, Mistral Large 4 56.4 from 5 benchmarks.
- Widest category
- Coding, 13.4 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.
| Category | MiniMax M3 | Bars | Mistral Large 4 | Ties |
|---|---|---|---|---|
| Coding | 2 | 12 | ||
| Intelligence | 0 | 2 | ||
| Agentic work | 2 | 4 | 1 tie | |
| Reasoning | 5 | 4 | ||
| Math | 1 | 3 | ||
| Knowledge | 4 | 1 | ||
| Web design | 1 | 8 | ||
| General | 12 | 12 | ||
| All shared boards | 27 | 46 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index212
| Benchmark | MiniMax M3 | Mistral Large 4 | Leader |
|---|---|---|---|
| LiveBench coding | 70.9% | 60.0% | MiniMax M3 |
| LMArena WebDev | 1482 | 1543 | Mistral Large 4 |
| AA Coding Index | 58.6 | Not measured | Only MiniMax M3 measured |
| LiveCodeBench (Vals AI) | 82.2% | Not measured | Only MiniMax M3 measured |
| SWE-bench Verified (Vals AI) | 75.0% | Not measured | Only 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 measured | 45.3% | Only Mistral Large 4 measured |
| Terminal-Bench 2.1 (Vals AI) | 53.6% | Not measured | Only 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 measured | Only MiniMax M3 measured |
| CyberBench (Vals AI) | 65.0% | Not measured | Only MiniMax M3 measured |
| SciCode | 47.1% | 54.2% | Mistral Large 4 |
| Terminal-Bench Hard (Artificial Analysis run) | 42.4% | Not measured | Only MiniMax M3 measured |
| FrontierCode | 14.7% | 39.4% | Mistral Large 4 |
| ALE-Bench | 640.0 | 1213.2 | Mistral Large 4 |
| LMArena Text: coding | 1470 | 1480 | Mistral Large 4 |
| LMArena Text: software and IT | 1466 | 1468 | Mistral Large 4 |
| LMArena WebDev: React | 1478 | 1558 | Mistral Large 4 |
| LMArena WebDev: HTML | 1454 | 1495 | Mistral Large 4 |
| LMArena WebDev: from an image | 1470 | Not measured | Only MiniMax M3 measured |
| Neither model has a result on the other 20 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | MiniMax M3 | Mistral Large 4 | Leader |
|---|---|---|---|
| AA Intelligence Index | 29.2 | 38.4 | Mistral Large 4 |
| Epoch Capabilities Index | 146.9 | 150.0 | Mistral Large 4 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index24
| Benchmark | MiniMax M3 | Mistral Large 4 | Leader |
|---|---|---|---|
| AA Agentic Index | 29.5 | Not measured | Only MiniMax M3 measured |
| Design Arena: agents building web apps | 1200 | Not measured | Only MiniMax M3 measured |
| Design Arena: agents building full-stack apps | 1197 | Not measured | Only MiniMax M3 measured |
| Design Arena: agents building mobile apps | 1169 | Not measured | Only MiniMax M3 measured |
| Design Arena: agents building Android apps | 1160 | Not measured | Only MiniMax M3 measured |
| Design Arena: agents building games | 1159 | Not measured | Only MiniMax M3 measured |
| Design Arena: HTML slides | 1086 | Not measured | Only MiniMax M3 measured |
| Design Arena: PowerPoint slides | 1169 | Not measured | Only MiniMax M3 measured |
| Design Arena: spreadsheets | 1042 | Not measured | Only MiniMax M3 measured |
| LiveBench research | 68.0% | 28.1% | MiniMax M3 |
| LiveBench browser | 8.0% | 8.0% | Tie |
| LiveBench long agentic | 12.4% | 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 measured | 17.5% | Only Mistral Large 4 measured |
| Excel Modeling Benchmark (Vals AI) | 47.8% | 56.2% | Mistral Large 4 |
| BioMysteryBench (Vals AI) | Not measured | 67.0% | Only Mistral Large 4 measured |
| τ²-Bench Telecom | 88.9% | Not measured | Only MiniMax M3 measured |
| APEX-Agents | 37.7% | Not measured | Only MiniMax M3 measured |
| OSWorld 2.0 | 4.6% | Not measured | Only MiniMax M3 measured |
| Vending-Bench 2 | $2,158 | Not measured | Only MiniMax M3 measured |
| Neither model has a result on the other 20 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index54
| Benchmark | MiniMax M3 | Mistral Large 4 | Leader |
|---|---|---|---|
| GPQA Diamond | 92.7% | Not measured | Only MiniMax M3 measured |
| LiveBench reasoning | 74.5% | 83.9% | Mistral Large 4 |
| MysteryMechanism (Vals AI) | Not measured | 12.2% | Only Mistral Large 4 measured |
| MMMU Pro (Vals AI) | 81.2% | Not measured | Only MiniMax M3 measured |
| GPQA Diamond (Artificial Analysis run) | 92.9% | Not measured | Only 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 |
| SimpleBench | 45.8% | Not measured | Only MiniMax M3 measured |
| Chess puzzles | 13.0% | 33.0% | Mistral Large 4 |
| Mystery game puzzles | 8.0% | 17.0% | Mistral Large 4 |
| LMArena Text: hard prompts | 1447 | 1446 | MiniMax M3 |
| LMArena Text: expert | 1461 | 1449 | MiniMax M3 |
| LMArena Text: hard prompts in English | 1455 | 1443 | MiniMax M3 |
| LMArena Vision: diagrams | 1266 | 1288 | Mistral Large 4 |
| LMArena Vision: homework | 1268 | Not measured | Only MiniMax M3 measured |
| Neither model has a result on the other 8 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index13
| Benchmark | MiniMax M3 | Mistral Large 4 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 71.1% | 98.3% | Mistral Large 4 |
| FrontierMath, tiers 1 to 3 | Not measured | 58.6% | Only Mistral Large 4 measured |
| FrontierMath, tier 4 | Not measured | 26.8% | Only Mistral Large 4 measured |
| ProofBench | 18.0% | 10.0% | MiniMax M3 |
| LMArena Text: math | 1429 | 1467 | Mistral Large 4 |
| LMArena Text: mathematical work | 1448 | 1467 | Mistral Large 4 |
| MathArena | Not measured | 26.7% | Only Mistral Large 4 measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index41
| Benchmark | MiniMax M3 | Mistral Large 4 | Leader |
|---|---|---|---|
| MMLU Pro | 84.2% | Not measured | Only MiniMax M3 measured |
| LegalBench (Vals AI) | 85.4% | Not measured | Only MiniMax M3 measured |
| CorpFin (Vals AI) | 68.1% | Not measured | Only MiniMax M3 measured |
| TaxEval (Vals AI) | 72.7% | Not measured | Only MiniMax M3 measured |
| MortgageTax (Vals AI) | 68.4% | Not measured | Only MiniMax M3 measured |
| MedCode (Vals AI) | 46.3% | 40.7% | MiniMax M3 |
| SimpleQA Verified | Not measured | 20.0% | Only Mistral Large 4 measured |
| LMArena Text: science | 1451 | 1445 | MiniMax M3 |
| LMArena Text: medicine and healthcare | 1439 | 1432 | MiniMax M3 |
| LMArena Text: legal and government | 1436 | 1427 | MiniMax M3 |
| LMArena Text: business and finance | 1434 | 1441 | Mistral Large 4 |
| Neither model has a result on the other 9 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index18
| Benchmark | MiniMax M3 | Mistral Large 4 | Leader |
|---|---|---|---|
| Design Arena: websites | 1257 | 1258 | Mistral Large 4 |
| Design Arena: UI components | 1243 | 1273 | Mistral Large 4 |
| Design Arena: game dev | 1211 | 1226 | Mistral Large 4 |
| Design Arena: data viz | 1236 | 1275 | Mistral Large 4 |
| Design Arena: 3D | 1209 | 1252 | Mistral Large 4 |
| Design Arena: all code categories | 1246 | 1259 | Mistral Large 4 |
| Design Arena: SVG | 1161 | 1232 | Mistral Large 4 |
| Design Arena: ASCII art | 1161 | 1202 | Mistral Large 4 |
| LiveBench design | 66.2% | 65.6% | MiniMax M3 |
GeneralLike-for-likePublic boards, shown as published, kept out of the index1212
| Benchmark | MiniMax M3 | Mistral Large 4 | Leader |
|---|---|---|---|
| LiveBench data analysis | 76.2% | 76.5% | Mistral Large 4 |
| Public Benefits Bench (Vals AI) | 64.1% | Not measured | Only MiniMax M3 measured |
| SAGE (Vals AI) | 50.6% | Not measured | Only MiniMax M3 measured |
| MedScribe (Vals AI) | 87.3% | 80.4% | MiniMax M3 |
| Vals Multimodal Index | 60.0% | Not measured | Only MiniMax M3 measured |
| IFBench | 82.9% | Not measured | Only MiniMax M3 measured |
| Long context reasoning (AA-LCR) | 83.0% | 81.3% | MiniMax M3 |
| DTBench | 78.9% | 88.2% | Mistral Large 4 |
| LMCA | 33.7% | 35.8% | Mistral Large 4 |
| Surface Evolver Bench | 55.0% | 28.1% | MiniMax M3 |
| Furniture assembly | Not measured | 20.8% | Only Mistral Large 4 measured |
| LMArena Text | 1432 | 1432 | Mistral Large 4 |
| LMArena Text: instruction following | 1432 | 1424 | MiniMax M3 |
| LMArena Text: creative writing | 1403 | 1364 | MiniMax M3 |
| LMArena Vision | 1251 | 1272 | Mistral Large 4 |
| LMArena Text: English | 1442 | 1433 | MiniMax M3 |
| LMArena Text: non-English | 1419 | 1423 | Mistral Large 4 |
| LMArena Text: Chinese | 1465 | 1494 | Mistral Large 4 |
| LMArena Text: German | 1430 | Not measured | Only MiniMax M3 measured |
| LMArena Text: Japanese | 1379 | 1411 | Mistral Large 4 |
| LMArena Text: Korean | 1372 | 1366 | MiniMax M3 |
| LMArena Text: French | 1443 | 1489 | Mistral Large 4 |
| LMArena Text: Spanish | 1432 | 1469 | Mistral Large 4 |
| LMArena Text: Russian | 1427 | 1418 | MiniMax M3 |
| LMArena Text: Polish | 1425 | Not measured | Only MiniMax M3 measured |
| LMArena Text: multi-turn | 1443 | 1441 | MiniMax M3 |
| LMArena Text: longer queries | 1445 | 1429 | MiniMax M3 |
| LMArena Text: writing and language | 1418 | 1387 | MiniMax M3 |
| LMArena Text: entertainment and media | 1402 | 1377 | MiniMax M3 |
| LMArena Vision: English | 1263 | 1283 | Mistral Large 4 |
| LMArena Vision: Chinese | 1311 | Not measured | Only MiniMax M3 measured |
| LMArena Vision: reading text | 1259 | 1284 | Mistral Large 4 |
| LMArena Vision: humor | 1235 | Not measured | Only MiniMax M3 measured |
| LMArena Vision: recognizing things | 1197 | Not measured | Only MiniMax M3 measured |
| LMArena Vision: creative writing | 1234 | Not measured | Only MiniMax M3 measured |
| LMArena Document | 1435 | Not measured | Only MiniMax M3 measured |
| EQ-Bench 4 | 1150 | Not measured | Only MiniMax M3 measured |
| Neither model has a result on the other 11 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | MiniMax M3 | Mistral Large 4 | Leader |
|---|---|---|---|
| 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 costs less
Repository review
50K input + 3K output tokens
MiniMax M3 costs less
Agent loop
200K input + 10K output tokens
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.
| Specification | MiniMax M3 | Mistral Large 4 |
|---|---|---|
| Maker | MiniMax | Mistral AI |
| Weights | Open weights | No public weights listed |
| Released | May 31, 2026 | October 6, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $0.30 | $1.36 |
| Output $/Mtok | $1.20 | $4.18 |
| Coding index | 39.2 | 56.4 |
| Independent measurements behind it | 6 | 5 |
| Cost per solved task | $0.56 | Not computable |
| Source | OpenRouter listing | OpenRouter 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?
74 benchmarks have a result for both models as of October 23, 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.